To get rid of the `Ident::empty` uses.

This requires introducing `PathParser::word_sym`, as an alternative to
`PathParser::word`.

By using `@` patterns more.

Also, use `Symbol` more in a couple of errors to avoid some unnecessary
conversions to strings. This even removes a lifetime.

…f they are provided

And note that the same limitation applies to all LLVM-based compilers

Co-authored-by: Ralf Jung <[email protected]>

… than linux

See <https://forge.rust-lang.org/triagebot/note.html>.

…onszelmann

Eliminate `word_and_empty` methods.

To remove the last remaining `Ident::empty` uses.

r? ``@jdonszelmann``

…ulacrum

Clarify black_box warning a bit

Trying to bring the docs on black_box more in line with the advice that we have discussed in Zulip.

rust-lang#140341 (comment)

… r=lcnr

Only include `dyn Trait<Assoc = ...>` associated type bounds for `Self: Sized` associated types if they are provided

Since rust-lang#136458, we began filtering out associated types with `Self: Sized` bounds when constructing the list of associated type bounds to put into our `dyn Trait` types. For example, given:

```rust
trait Trait {
    type Assoc where Self: Sized;
}
```

After rust-lang#136458, even if a user writes `dyn Trait<Assoc = ()>`, the lowered ty would have an empty projection list, and thus be equivalent to `dyn Trait`. However, this has the side effect of no longer constraining any types in the RHS of `Assoc = ...`, not implying any WF implied bounds, and not requiring that they hold when unsizing.

After this PR, we include these bounds, but (still) do not require that they are provided. If the are not provided, they are skipped from the projections list.

This results in `dyn Trait` types that have differing numbers of projection bounds. This will lead to re-introducing type mismatches e.g. between `dyn Trait` and `dyn Trait<Assoc = ()>`. However, this is expected and doesn't suffer from any of the deduplication unsoundness from before rust-lang#136458.

We may want to begin to ignore thse bounds in the future by bumping `unused_associated_type_bounds` to an FCW. I don't want to tangle that up into the fix that was originally intended in rust-lang#136458, so I'm doing a "fix-forward" in this PR and deferring thinking about this for the future.

Fixes rust-lang#140645

r? lcnr

…orm, r=lcnr

Structurally normalize in range pattern checking in HIR typeck

Fixes rust-lang/trait-system-refactor-initiative#200

r? lcnr

…=jieyouxu

Improve `-Zremap-path-scope` tests with dependency

This PR greatly improves our coverage of `-Zremap-path-scope` for diagnostic paths and macros with dependencies.

r? ``@jieyouxu`` (since we talked about it)

try-job: x86_64-msvc-1

…oval, r=jieyouxu

Make `rustdoc-tempdir-removal` run-make tests work on other platforms than linux

Follow-up of rust-lang#140706.

r? ``@jieyouxu``

…lbini

Add release notes for 1.87.0

Originally drafted in rust-lang#140133

cc ``@rust-lang/release``

r? ``@pietroalbani`` as you're running the release

Enable triagebot note functionality for rust-lang/rust

Docs: <https://forge.rust-lang.org/triagebot/note.html>.

r? ``@Urgau``