It seems to me there are a two main reasons to believe a theorem/conjecture to be true:
Because it has a correct proof (e.g. the Feit-Thompson Theorem, Dirichlet's Theorem)
Because there is an intuitive reason/heuristic for it to be true (e.g. $P \ne NP$, Dirichlet's Theorem)
The former is of course usually the gold standard. However very long proofs may still be regarded as suspicious after a proof has been produced. The longer a proof is, the more likely it is to have undiscovered errors, so that even if no specific gap is known, people don't trust that there isn't a gap. Computer verification of proofs may be used to alleviate these concerns, e.g. the Feit-Thompson theorem was verified with Coq in 2012.
But it seems neither condition convincingly applies to CFSG.
It has a proof, but this proof is tens of thousands of pages spread across hundreds of articles, and the second generation is still going to be thousands of pages long. Moreover, at least one mistake have previously been discovered in the proof: the gap filled by Harada and Solomon in 2008. I am aware CFSG has likely been submitted to significantly more scrutiny than the average proof, but even so, it seems plausible (from my uninformed perspective) that an error could have slipped through. And CFSG has not been computer verified.
There also seems to be no intuitive reason to expect the theorem to be true, as this MathOverflow question suggests. Apparently it wasn't known if there were finitely many sporadic groups until late in the proof, suggesting even to those familiar with the proof there is nothing intuitive about it. Though maybe there is something intuitive about it to experts now.
However, the theorem is widely trusted by experts, so I would like to to understand why.
What details I am interested in depends on what the answer is, but I am not asking about the structure of the proof itself. My impression is that it is widely believed there may be mistakes in the proof, but any mistake would be fixable. I don't understand how one can be confident that unfixable mistakes don't exist without being confident that no mistakes exist, unless there is an intuitive reason the theorem must be true. So if my impression is correct, an explanation of how you can be confident about one without the other would probably be satisfactory. This might not be anything specific to CFSG.
$$ $$
So my first question is: Am I correct that it is widely believed there may be errors in the proof of CFSG?
If so, how can we be confident those mistakes are fixable? Is there something intuitive about the proof itself that suggests it should be true? (if there is, I accept that I won't be able to understand what the intuitive thing is)
If not, how can we be confident there are no errors without computer verification?
This is not a duplicate of How confident can we be about the validity of the classification of finite simple groups? as that confirms that we are confident in the result, but doesn't explain where this confidence comes from.
I do not think this question is opinion based, as I am asking why experts trust it, rather than if the person responding trusts it.