Players are often stuck because they simply don't know what they should do. Percival's message deals with player engagement very well, but I'd like to offer a slightly different take on this question.
If you can't figure out how the plot might be solved, players definitely won't
Players' ingenuity often amazes GMs, but you can't assume they will figure out the mystery for you. You have to leave them clues and hints pointing them in the right direction. This is especially true in non-railroading games. Think of it this way: if the story is about them tackling highwaymen but there are many paths they can take, you need a lot of signs and pointers to guide them to the destination.
For every mystery, establish three solutions, for every solution, three clues.
That is apart from what the players can invent themselves. There is an excellent article describing in detail why and how you should do it. In general, the players are often overwhelmed in your world - they are not Sherlock Holmes. You need to give them at least three ways to accomplish the narrative goal and for each of those ways, drop three hints that they can gather.
Be open to their own solutions
The players might decide to do something really innovative, like going to a clairvoyant hermit for advice or using bribery to have people rat the true criminals out. It should work and give them at least a clue if not be the solution outright. Unless, of course, it's completely inappropriate, in which case it should fail, but still give them a clue. Reward players' actions with clues, regardless whether they are right or wrong.
The clues have to be obvious
Fencepost to the head levels of obvious. It's not enough that they find a curious dagger, it has to be a dagger adorned with Thieves' Guild symbols, with Thieves' Guild telltale ornaments and screaming Thieves' Guild. Otherwise instead of a clue you present another mystery. You don't want a player saying "Oh, we have the dagger but what does it mean?"
No red herrings
It feels awful to pursue a goal and then discover you're at square one. The players could be hindered, but if you throw their theory out of the window it will be nonetheless frustrating. At least part of the theory should be relevant, so maybe the Guild is not the main criminal, but they helped them?
It all comes down to agency
Your players need to be able to take action. If they don't know what they should do, you need to give them more information. Introduce characters that will help them know how to deal with criminals, what are the success criteria. How about a guardsman who will tell the heroes "Well, the Thieves must know who did it, if they didn't do it themselves." Don't you think the players would act differently (and decisively) if they heard that? More on agency can be found here.
Solving your case
OK, so let's apply these to your adventure. Establishing the facts:
The robberies have not been done by the Guild, but by the Cult of Shadow. Of course, you can insert any villain instead of the Cult.
To succeed, they need to find out that it was the Cult of Shadow.
Three solutions:
- Set a trap on the trade route or accompany a caravan.
- Investigate the robbed caravans, find out robbers are after holy symbols.
- Find the Cult's hideout full of stolen wares.
Now, the clues:
1a The Thieves Guild will happily set a trap if it takes the Sheriff's eyes off them.
1b The woods have a certain place where all robberies seem to happen.
1c The merchants are hiring mercenaries as protection for next caravans.
2a The local temple posted a reward for robbed holy symbol.
2b Thieves Guild claims the caravans robbed carried only worthless religious artifacts
2c The remains of a caravan show a left behind reliquary.
3a A priest complains about ghosts in the abandoned mine, where supposedly lights shine in the night.
3b One Thief is found dead, his blood tracks lead to the hideout.
3c The Sheriff says there are masked men witnessed in the hideout area, but doesn't think it's important at the moment.
With this amount of clues, it's unlikely they would get stuck.
The rules don't provide any way for the average character to do this.
While it's theoretically possible to jump up to the altitude of a flying monster, one would need to be able to stop in mid-air for long enough to make their attack, and without some outside help (e.g. some form of flight, tree branches to cling to, etc), characters just can't really do that. A jumping Charge also wouldn't work, since the character would still be unable to end its movement close enough to the enemy to attack it.
If the monsters lacked any ranged attacks themselves, the melee PCs might consider readying actions to attack the flying enemies when they come in range to attack. Otherwise, the situation ends up being a hard lesson in melee characters always bringing a ranged option (hopefully the DM goes easy on them the first time).
Of course, as usual, the DM can always adjudicate their own solution with a house rule, like you did. Seems reasonable enough. Though I'm not sure what good the attempt would do in the first place, considering you'd need a DC 50 Athletics check just to get 5 feet off the ground with a high jump, or 25 if you can somehow get a running start with your charge.
Regarding the question of a jumping Charge:
The rules provide no more support for a charge using a high jump to reach a flying enemy than any other form of jump attack. Despite being a single Standard action involving both moving and attacking, the two portions are separate and distinct. The move must be completed before the attack can be made, just like any other power that allows you to move and then attack without explicitly stating you can attack in the middle of the move, and so you must be able to stop your movement in the final square of the charge.
Ending the move in mid-air leaves you in an invalid space, whose consequence is falling, which thus happens before the attack portion and occurs effectively instantaneously (assuming high-altitude fall rules aren't relevant to the situation). If it were possible to charge into mid-air and make an attack before falling, it would also be possible to make a charge ending in an ally's space and attack before the consequences of ending in an invalid space occurred (in this case, forcing you back to the last valid space you occupied). It should be clear that this is not intended as a possible use of a charge attack.
Another less-obvious side effect of assuming the rules allow a jumping charge is that falling does not provoke opportunity attacks, unlike normal movement, and thus the charging creature would be able to retreat from the enemy without consequences after completing its attack.
Best Answer
From page 74 of the 5e Basic Rules:
Or in the PHB p.195:
So yes, you can grapple someone and then drag them. Now in this case the real unanswered rules question is that you are dealing with an "unopposed grapple" - the wizard isn't looking to avoid it, so does the barbarian need to roll to hit or what? I normally allow such a move to succeed if unopposed, unless it's the prototypical "guy is falling down a pit/off a wagon/etc" situation where there's a question of whether the barbarian could easily lay his hands on the wizard in the first place.
Keep in mind they'd both be grappled, and the barbarian possibly encumbered, while doing all this, though you can disengage a grapple at will. But other than that, sounds like you handled it fine. The barbarian could probably have mounted his horse still carrying the wizard along, as he could also lift him (see the push, drag, and lift rules under Strength, you can do all three to the same weight for a given Strength). Actually getting the wizard to mount another horse would probably incur the same "half movement" cost as mounting it himself.