[SMTChecker] Refactor pointers

[leonardoalt]

There are a lot of shared_ptr being used for smt::Sort where they could be raw pointers.

[leonardoalt]

@ekpyron actually why did we want them to be raw pointers instead of shared_ptr?

[ekpyron]

shared pointer have non-zero overhead and you always need to be careful about them becoming cyclically dependent on each other...
In general, it's always nicer to have clearly defined explicit ownership and lifetimes instead - but not sure if there was any more concrete reason than that...

[leonardoalt]

Yea, tho I'm wondering if that really makes sense for the SMT sorts

[ekpyron]

Rather similar case to us introducing that TypeProvider thing for the solidity AST I guess.

[aniketmdinde]

I want to contribute to solve this issue. Can you please guide and assign me this issue?

[DavidRomanovizc]

Hello, @leonardoalt this problem is still relevant?

[Ajay-26]

Why not use unique_ptrs as opposed to raw pointers ? Also, I would like to contribute to this issue if possible

[Pawar7349]

I’m interested in working on this issue of refactoring shared_ptr to raw pointers or unique_ptr. Could you provide some guidance on the preferred approach?