The solution is for the PCs to experiment with each of the spells to get a better feel for how useful they are during an actual game. Once your players have a chance to experiment a bit with the spells they have available to choose from, the task of deciding which ones are worth picking will be significantly easier. Also, as the DM, you may wish to reward good Intelligence or Wisdom checks made by the PCs during research activities by giving them suggestions as to which spells might prove useful over the course of the adventure. For example, if you're running the Lost Mine of Phandelver, you may wish to reward a player who spends time researching dragons that Green Dragons are masters of poison, so a Protection from Poison spell may be worth the Cleric's time to have prepared.
Because many of the 5th Edition mechanics and spell changes will likely be new for many players who did not participate in the playtest, the Adventurer's League Player's Guide also offers a nice solution to this problem. During levels 1-4 (the first major tier of gameplay) you can rebuild your character at the end of an episode or an adventure. This will allow you to try out low-level spell combinations without fear of being stuck with poor choices. Encourage low-level players to pick spells that sound cool and appear useful but not to fret over making bad choices because at the end of the session they can retcon their choices anyway.
It's worth mentioning that this should be significantly less of a problem for the Cleric, Druid, or Paladin because they each automatically have access to their entire class's spell list. If they wish to change the spells they have prepared, they can do so at the end of a long rest.
Lastly, note that some spells (those with the ritual tag) can be cast as rituals. This means the casting time is increased by 10 minutes (to prevent them from being used in combat), but the spell doesn't use up a spell slot. Bards can cast any ritual that they know. Clerics and Druids must prepare their rituals just like any other spell they plan to cast, but Wizards can use their spellbooks to avoid preparing rituals ahead of time. Warlocks who take Pact of the Tome have an invocation option which gives them the ability to learn every class's rituals in the same way as a Wizard. There is also a feat (Ritual Caster) which can grant ritual spellcasting to the other classes. This too will allow your players to try out more utility spells that may or may not be useful without making them sacrifice more pragmatic options like combat spells.
I know you're not a native English speaker, but in many circles of general nerd culture in America and on the Internet, there is a phrase for what you're doing right now: your spaghetti is falling out of your pockets.
In all seriousness, though, you seem very anxious about your situation and you should take a step back and relax. You say you're good at improv, but it looks like you're already sweating bullets because your party got sidetracked instead of getting to the quest you wanted to take them on.
I'm going to put my direct answer to your question first and some general advice afterwards.
Your players probably won't miss their characters. Why? Well, you generated them, and even wrote up their backstories. These characters were never made by the players playing them, and while they might have grown attached over a few moments, I find it extremely unlikely that they'll shed tears over a party that lasted for four sessions... that they didn't even make themselves. Even if the characters are extremely cool and interesting, that element of personal attachment just isn't there, which allows the players to truly say that they helped create an exciting story with interesting characters. This obviously varies from person to person, but again, I find it very unlikely that your players would have grown very attached in this case.
Go out with a bang. You know that amazing epic encounter you were saving for the climax of the plot arc? Yeah, run it now. You'll have to make some tweaks because you obviously aren't quite there yet, but get them there as fast as possible. Feel free to kill off PCs or even have a TPK at this point; memorable deaths are often much better than "and then they lived happily ever after."
Here's a few things to keep in mind when starting your next campaign:
Your content will come to light eventually, and it will be good. The quests you've designed will always find a way to come forward. Even if your plot arc is entirely ruined by something the players did, you will be able to recycle the content you made but never ended up playing, and I encourage you to do so for your new campaign. The only things that are truly lost are "hard" materials, like NPC stat sheets, etc.
You're in control, and therefore, you set the tone. Sometimes, it is best to take a page out of Gygax's book; after all, this is your campaign, and you put a lot of effort into it. Obviously you shouldn't take the entire preface from the AD&D DM's Guide to heart, but there is a point where a DM should draw a line in order for there to be some kind of structure, assuming you want your campaign to go anywhere. If your players are goofing around and killing NPCs for no reason, or making light of important people in-character, then they should be ICly punished for it; reprimanded for insulting a nobleman, pursued for attacking innocents, etc. It is also very possible to play a serious game in character and laugh until you're blue in the face out of character. This frequently occurs in the Dark Heresy games that I've played and ran.
You had better get used to murdering your darlings. This is a phrase commonly used amongst writers and creative designers everywhere in the U.S. The phrase means that you'll have to scrap ideas frequently, including ones that you really, really liked, so you had better get used to it. The saying is intended for use in the writing, film, video game, and other industries where a publisher or producer oversees your work, constantly telling you what can and can't make it to the final product based on time and expenses. However, it works just as well for when your ideas can't make it to the game because your players did something insane. And, on a related note...
Plan less. I don't know how much effort you're putting into writing everything ahead of time now, but you might want to ease up on that. From what you're telling me about your role-playing experience, it seems like you've been playing in a "safe" and slow environment where you rarely, if ever, have to scrap or re-do material. This happens literally all the time in regular tabletop RPGs, thanks to the insanely unpredictable nature of 4-5 different people working together. It will save you a lot of anguish if you lay out a basic outline of what's going to happen and then add the details once you're sure the players will be arriving there next session, or maybe two sessions later.
It seems like you've learned a lot already OP, which is great, but scrapping a campaign after 4 sessions (and while your players are all enjoying it) is something you should really avoid. If everyone else is having fun, consider either shaking things up a bit and changing your own notes, or coming up with a way to set them back on track, which doesn't always need to feel contrived or railroad-y.
EDIT: Well, now that SevenSidedDie has made that edit to your post, there are a couple of details that I didn't quite catch before, no offense. Since you said you have a month between each session, it seems like you might be over-planning because you have a lot of time between sessions. Heck, you might even consider having more frequent sessions, if you can't stop yourself from overthinking it in the intervening months. If in-person is not an option, use Skype and/or Roll20.
Best Answer
There's an apparent paradox in character creation for an unfamiliar game: to effectively and confidently make a character requires knowing the game, but to know the game you have to already have made and played a character. It's not really a paradox, but it can feel like it when you have limited time to play and want to get started as soon as possible in order to get right into the fun of an ongoing campaign.
There's a simple solution that has worked well and consistently for me, but which requires trusting that patience pays off.
Play a demo session first
What has been successful for me across RPGs is to run a one-shot demo session using the new game before we make our real characters in our real campaign. This introduces the players to the moving parts of a character through hands-on playing experience, which gives them a basic understanding of what's important when making characters. This can also effectively introduce the players to the setting and playstyle that they will be making characters for later.
Whether the demo characters are pre-made or made by the players doesn't seem to matter. (If they know these are throw-away characters, they don't suffer nearly as much analysis paralysis in making their own.) What matters most is that the demo session gives them an experience that reflects the realities of play that should be informing their character creation choices.
For example, in a game where understanding the skill system is critical for character creation and evaluating character effectiveness (RuneQuest 6), I have run an "obstacle course" session where they made characters and then played through an in-setting coming-of-age trial that involved a lot of skill use (but no combat or risk of death). By the end, players had a visceral understanding of what was and wasn't a good skill rating — one player initially thought that 35% was a good skill and spread their points around to hit that number in as many skills as possible, and came out of the demo realising that she'd underestimated that by half and that choosing a few core skills to maximise first, before spreading the other points around, was key. They all also profited from the crash course in the cultural context they'd later be playing in.
In another game where the interplay of character creation choices and combat is a big deal (Savage Worlds), I had them make one-off characters and then threw them into a dungeon that I knew well enough to run on-the-fly. They had the freedom to go where they wished and test their characters in a variety of non-combat and combat situations. As a result, they got a good sense of how the game functions overall and in its combat, magic, opposed skill, and healing subsystems in a very short time, and were confident making characters for the longer-scale fantasy campaign we later kicked off. Notably, when we started that campaign we had a new player, who had a much harder time creating her character than the ones who had the demo session initiation.
In both these examples, taking the time to give the group early hands-on experience with characters and the system meant that the players were confident and quick in future character creation for the real game. The difference was like night and day: where before they were lost and stumbling through the options, afterwards they were focused and dove into the chargen process with clear goals in mind.