To me it sounds like your player is a mix of being impulsive and a newbie to roleplaying. The newbie elements (needing stuff explicitly explained and such) should work themselves out with time. The impulsiveness usually needs a little bit of work.
Here's what I did once to rebuff the impulsive players in my campaign:
- Set up a wonderful campaign arc that involves the "imminent" death of the party.
- Put impulsive character in a situation where he finds the important "cure"/MacGuffin.
- Both the MacGuffin and dilemma turn out to be fake.
I did this to deal with a combat monster who kept doing stupid things (shooting up places, making a "peace gesture" that got the party shot at, and spooking an extraction target [leading to his death]), and it worked pretty well- he was the face of the group, and decided that when the "nanite antidote" that was supposed to cure the party's impending doom nanite fun death he would drink the whole thing to ensure he didn't die (because only two party members were actually really in danger of death, him being one). Turns out that there were no nanites, and the antidote was cyanide.
That was the opposite of what I should have done.
Mind you, it didn't destroy my group, the player stopped being impulsive, and life went on (for all but that one guy's character). But it was a stupid, brash, inexperienced GM maneuver, and it could've cost me a player. It did have the upside of making everyone else more paranoid, but the impulsive people are still impulsive, they just put a layer of paranoia on their actions (which I guess makes them less impulsive by definition, but doesn't promote good decision making).
Ultimately, you will run into these brash and (frankly) obnoxious players. It's not even a personal fault in them; the three or so I have/had (as their various states of rehabilitation qualify them) in my group are all really nice guys, but they just don't create a coherent character. So here's what I've started to do with them:
- Common Sense; Shadowrunners will recognize this as a name of an edge, and it basically reads like this: "Are you sure?". I encourage my players to all take this edge for all but the most oddball of characters (usually not an issue, since they tend to be played by the players who can handle themselves well).
- "The Talk"; tell them it has to stop, plainly and explicitly. I actually had to do this with one of my players (the cyanide one, in case anyone was wondering) when they played a crazy sociopath Malkavian in Vampire: The Masquerade. Go to the player and tell them in explicit language that their characters' actions have to stop. No qualifications, no debate. If not, character goes bye-bye entirely, due to the fact that he [insert appropriate gaffe here] (the Malkavian was on the intersection of "Shot up the First National Bank at dawn while wearing a Speedo" and "Built a functional nuclear bomb, fumbled while stashing it away", with the former being slightly more likely).
- Veto; most games include very prominent "ask your GM" clauses during character creation. Call that in. The player will fuss about it. They may leave. If they are that disruptive to the game and the group, however, it may be a necessary evil to tell them that their character cannot a) remain under their control and b) remain in the campaign. It doesn't necessarily mean that the character vanishes from the universe and never existed, but he's a NPC now, retires suddenly, or goes out in a blaze of glory. He does not, however, continue acting as he has and sticking with the group.
Disclaimer: You may have other factors leading to this issue.
Wrong Game: The player isn't actually interested in playing this game; even if they're interested in the setting and mechanics, they don't want to abide by them. This is what I call the "Sparkle Vampire" syndrome I occasionally have to deal with from a player who read all the supplements and got a bunch of ideas ("But the book says cyberzombies are only really, really, really hard to create!) that they then assumed would apply to their characters. These are the sort of people who want to play sentient variants of high-level D&D Monster Manual entries, fully sapient human-form mind animals, and the like. If they were playing Eclipse Phase they'd go for the Octomorph and give it a fancy cybernetic suite including jet thrusters.
Bad Player: I hesitate to call someone a "Bad Player", but it's true that some people prefer to play things revolving around them. While this is natural, some people take this to an additional extreme, and must make everything they play revolve around them all the time. Sometimes this leads to "The Talk" (see above), and sometimes this just means they won't have fun in the game and should pursue something else.
GM Ineptitude: Note that I'm not accusing you here, and I'll keep the examples my own. I used to run an Eclipse Phase game, and I made the players into hard hitting immortal cyborg soldiers. It lasted three runs. My players got bored because they had no consequences for failure. I ran a Remnants game. It failed because the players kept running into issues where their (overly large) group kept falling apart on matters of dogma or running into massively high power gradients. This same group has been in a Shadowrun campaign that lasted for almost a third of a year with weekly sessions, and the reason it ended was due to scheduling conflicts and getting far outside the realm of mortal power. It's not that this even means I'm a bad GM, it just meant that I was aiming for something and my players weren't, and there was a communication breakdown or I tried to push it on them too hard (or I over-hyped them).
Your first question is about half a GM problem, and half a player/PC problem. I've been on both sides of this issue: GMing for a character who had little reason to participate in the story with the other PCs, and playing a character who had no reason to participate in the story and every reason to run wildly in the other direction. In both situations, the solution has two parts:
The GM finds ways to include the PC in the plot.
It sounds like your group is already doing this, but for posterity: this means that the GM must look for ways to actively involve the PC in the story, by using their background hooks, talking to the player about what the character might find interesting, and otherwise looking for ways to help the character find motivation within the story to participate in the story.
You do need to be careful not to change the story or the game's focus so much that the other players begin to feel excluded or ignored, but this is a matter of understanding your group, finding a balance, and getting buy-in from the player of the problem PC. Which brings me to part 2 of the solution:
The player finds ways to include their character in the plot.
This is the much harder part of the solution: the player must meet the GM halfway, otherwise the game becomes "The Problem PC Show!" and no one else has fun. This may mean the player has to bend or tweak the character in some way - not enough to compromise the core of the character, but enough to keep them with the group when otherwise they wouldn't stay.
This is, admittedly, difficult to do. Some people are so attached to their characters that they're unwilling to compromise; or they simply can't see a way to bend the character on an issue without breaking them completely. However, it's absolutely required in order to solve this problem.
Why player buy-in makes a difference:
When I played a character who had no reason to participate in the story, I looked hard for ways to make her want to be there. But all logic dictated that she run far away, find a hole, and pull it in after herself, so any other choices felt wrong to me. This showed in my roleplaying, and ultimately caused much frustration for our whole group as session after session became centered around getting her involved when she didn't "want" to be.
I put "want" in quotes there for a reason. Players with especially strong "my guy" syndrome (which I know I'm prone to) will insist that they're "just playing their character" and that "the character can't be changed". But when I GMed for a character who had no reason to be there, his player was willing to meet me halfway, and the result was a fun campaign for everyone where that character even made a large part of the highlight reel.
The reason it worked was because the player was willing to say, "he has no reason to come along, but does anyway, because that's how group-based RPGs work." We hand-waved it a bit as "he has nowhere else to go and nothing better to do", but really that character should have been as long-gone as mine wanted to be. But because the player was willing to take the meta option and bend his character enough to say "screw it, he's participating" without looking for a story reason, it worked. It kept the game from focusing too much on that character, but meant he was still there when story developments happened that he was interested in, thus giving him the time he needed to organically grow interested in the plot.
When the player isn't interested in the scenario
This part is a little trickier. If the player doesn't buy Dresden's world, including all the magic elements, what's he doing playing a Dresden Files game anyway? You and your co-GM need to talk to this guy privately, out of game, and ask him why he's playing. If he just wants to hang out with your group and doesn't care much about the game, then you need to address that. Maybe give him a character with a minor support role, so that he doesn't need to participate much in the game and can just hang out; maybe say that if he just wants to hang out, the game isn't the time for it and you'll find other group activities he can participate in. It's up to you and your group to decide what's best here.
If he insists that he wants to play the game, then you need to find out why he's actively sabotaging it. Tell him that his actions suggest he doesn't want to play, and in fact are making it harder to play. It's possible you simply have an attention hog on your hands, in which case you should deal with him appropriately. (I'd suggest going through the problem-players tag, as there are a number of questions and answers you may find relevant depending on your exact situation.)
TL;DR: The GM and the player need to meet halfway on adapting the character and the world. But if the player isn't buying the game's scenario in the first place, then that must be addressed first.
Best Answer
The goal is to build an incentive system where "I'm going to farm for a while" yields less XP than "we're all going to advance the adventure together". That way, adventuring together is the most effective way to get more powerful.
Step one, disincentivize solo farming. Every build has a counter. He's stealthy? Monster with blindsight. Wizard with alarm's throughout the area. Divinations. The ultimate sentry, who put so much into perception and investigation, and has a background of Sentry (you'll have to make that background) that, within their limited area of expertise, they are a god; they have clear-cut the area, burned the undergrowth to the ground, and hid 1,000 pit traps around their camp.
Know what's really boring/doesn't give you a lot of XP? When you fall in a pit and break your leg, and have to wait for your party to rescue you (accumulating XP as they go).
Immobilization traps can be great for this. They won't straight up kill him, but the humiliation of being repeatedly rescued by the rest of the party (who are accumulating XP in the process) should very quickly wear him down.
Adventurers form parties because adventuring is dangerous. If this guy is able to cut a swath through your encounters solo, it's because you need to up the challenge of those encounters.
On top of that, life doesn't stop for the rest of the party while he's gallivanting around. Maybe they stumble across some treasure. Or get ambushed. If they get ambushed by something that would have been really useful to have that rogue to face and they waste a lot of resources unnecessarily, then the others will start pushing the problem player to stick with the group.
Every time he says he's going off on his own, I'd say, "Okay, you go scout through the forest, yadda yadda yadda, we'll come back to that. Everyone else..." Anyone who aggressively steals the spotlight should be delayed before they get that spotlight.
Personally, I don't like messing with the XP system too much (as others are suggesting). By and large in my games, people in a fight get XP, even if, really, the fighter just cut through everything. You get the same XP for avoiding a fight (stealth should not get penalized). You get half XP for being forced to retreat from a fight (you didn't win, but you did learn something). There are also story rewards: you found the lost treasure of [whatever], rescued the mayor, whatever. Finally, every player gets "bonus" XP for being particularly effective at a crucial moment, roleplaying well, overcoming a personal challenge or weakness, or, and this one's key, increasing party cohesion.
Finally, the carrot: build encounters around this guy. He wants to stealth around? Have him scout along a cliff, taking out archers that are threatening the group below. Everyone else is engaged in their own combat encounter with [insert whatever], everyone's on the same initiative, everyone's playing, everyone's getting XP. And make it clear he's getting more XP than he would on his own, since he's getting XP for the rest of the party's combat encounter (since this is really all just one combat encounter).
The only real problem I could see with a guy like this: he doesn't just want to be powerful, he wants to be more powerful than the other PCs. If that's the case, you're going to have to have a carefully worded talk with him. And if he's still a problem... you can either kill him until he gets the idea (and institute XP penalties for death, either on resurrection or on new character creation), or kick him out of the group.