Fix next-solver ICE on PointeeSized goals

[TaKO8Ki]

Fixes #151957

[rustbot]

r? @mati865

rustbot has assigned @mati865.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[mati865]

@rustbot reroll

[TaKO8Ki]

r? compiler

[jieyouxu]

r? types

[theemathas]

I think this fixes the wrong problem, since the ICE can also occur outside the next-solver. See #151957 (comment)

[rperier]

We should not reach this point, self.predicate_must_hold_modulo_regions(...) should not being called. After investigation, there is a false positive match between PointeeSized from the current crate and core::marker::PointeeSized . The problem being, we could have exactly the same issue with some other traits in core::marker::* (like Unsize . I mean some false positives matches without ICE)

[rust-bors]

☔ The latest upstream changes (presumably #152965) made this pull request unmergeable. Please resolve the merge conflicts.