feat: visibility for impl functions

[asterite]

Description

Problem

Part of #4515

Summary

After searching for a while I found that to make this work we only needed to change one line

Additional Context

I didn't include tests for now, but I'll add them later because we first need to figure out which impl functions should be marked as pub or pub(crate) in the stdlib. I'm not familiar with the stdlib though I probably can't do it. That said, if you open the standard library with a nargo compiled from this PR you already get some warnings, so those functions should at least not be private.

Documentation

Check one:

• No documentation needed.
• Documentation included in this PR.
• [For Experimental Features] Documentation to be submitted in a separate PR.

PR Checklist

• I have tested the changes locally.
• I have formatted the changes with Prettier and/or cargo fmt on default settings.

[TomAFrench]

turbofish_numeric_generic_nested_call is failing so it seems like private impl functions cannot be called from the same module (i.e. the functions are private to within the impl rather than to the module in which the impl sits)

[asterite]

Ah, right. I just bumped into this while trying to add some visibilities. It seems that also U128::method fails if you are inside U128, but it doesn't fail if you do Self::method 🤔 I'll take a look 😄

[asterite]

Done! All tests pass now. I think there are still some fns that could be made pub.

[TomAFrench]

This seems like a slightly different case to what I meant in my comment. I've pushed a new test case which I would expect to pass but doesn't.

[asterite]

I think I got it now! 😄

[asterite]

I noticed that visibility isn't checked for the new TypePath, which is something like Field::from_le_bytes where what's before :: isn't a Path but a Type. That requires different code to do it so I'd prefer to do it in a separate PR.

[TomAFrench]

Yeah, there's a good amount of stuff broken with type aliases due to this, a lot of issues related to type aliases stem from this #4469

[asterite]

I see.

Well, this is ready for review in case you want to take a look at it.

[TomAFrench]

One concern about a function we're making public but otherwise looks fine.