First commit is not about warnings, but instead addressing TODO about whether uint8_t is necessary

I've updated hint code to use size_t in places where the indexes are just being passed around, & not actually stored anywhere. At best uint8_t in these scenarios generates similar code, at worst you have the compiler generating unnecessary code to maintain 8 bit semantics on size_t registers

this seemed to fix something for @Pepper0ni

You've marked a lot of classes final, and there's nothing wrong with that. But, the root of the problem is that their parent classes do not have a virtual destructor. These warnings are going to crop up again if more class inherit from these parents. These problems should be fixed upstream, like I've done in Kenix3/libultraship#871. I see you've address this issue in Kenix3/libultraship#886, but I'll stress that the better fix is to mark the parents' destructors as virtual. It is good practice in C++ to have virtual destructors if a class is intended to be inherited from, and expecting children to be final will lead to obscure bugs.