fix: remove alias rigid before calling relate

[cordx56]

This PR fixes #158773.
This fix resolves regression from at least nightly 1.96.0, as I checked.

Cause of issue

In the current code, relate(a, b) is called on the type SameTypeModuloInfer, which is outside of the trait resolver.
So the rigid should be removed.

I inserted ty::set_aliases_to_non_rigid before calling relate method and now it works correctly.

[rustbot]

Thanks for the pull request, and welcome! The Rust Project is excited to review your changes, and you should hear from @mati865 (or someone else) some time within the next two weeks.

Please see the contribution instructions for more information. Namely, in order to ensure the minimum review times lag, PR authors and assigned reviewers should ensure that the review label (S-waiting-on-review and S-waiting-on-author) stays updated, invoking these commands when appropriate:

• @rustbot author: the review is finished, PR author should check the comments and take action accordingly
• @rustbot review: the author is ready for a review, this PR will be queued again in the reviewer's queue

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler
• compiler expanded to 75 candidates
• Random selection from 22 candidates

[cordx56]

Hi, @adwinwhite. I found the issue is easily resolved and fixed in this PR.

If you already working on this problem, sorry for taking it without your confirm.
And if you find any problems, I would appreciate any comments.

[adwinwhite]

Hi, could you check whether the problem can be fixed at the root? Like if there's something that should be normalized but not. set_aliases_to_non_rigid is kinda a hack. See the comments of it.

[cordx56]

Thanks for reviewing! Okay I will check it. But I will do it after my working hours so it will be late…

[rustbot]

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

[cordx56]

Hi, I pushed new code that does not use set_aliases_to_non_rigid.
I'm not confident that the code fixes the problem at the root, but it may be a large task if we explore other fixes that capture the root cause.

[mati865]

r? @adwinwhite

[adwinwhite]

Sorry for being so slow. Could you try deeply_normalize_for_diagnostics before relating in same_type_modulo_infer?