Removed in_band_lifetimes from rustc_trait_selection

[lameferret]

Issue: #91867

Wasn't able to run ./x.py build --stage 0 compiler/rustc_trait_selection/ got
thread 'main' panicked at 'error: no rules matched compiler/rustc_trait_selection/', src/bootstrap/builder.rs:236:17 only ran ./x.py check
Is this the supposed behaviour ?

[rust-highfive]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @oli-obk (or someone else) soon.

Please see the contribution instructions for more information.

[oli-obk]

Is this the supposed behaviour ?

Well... kind of. Even if that one crate compiles, you still need to check if the rest of the compiler still works, as other parts of the compiler depend on that one crate.

@bors r+ rollup

[bors]

📌 Commit 486ee57 has been approved by oli-obk

[jackh726]

Duplicate of #91904, sorry!

[camelid]

Wasn't able to run ./x.py build --stage 0 compiler/rustc_trait_selection/

Why are you passing --stage 0? Don't pass a --stage flag in this case.

[lameferret]

Wasn't able to run ./x.py build --stage 0 compiler/rustc_trait_selection/

Why are you passing --stage 0? Don't pass a --stage flag in this case.

To build a compiler, run ./x.py build. This will build up to the stage1 compiler, including rustdoc, producing a usable compiler toolchain from the source code you have checked out.
For hacking, often building the stage 1 compiler is enough, but for final testing and release, the stage 2 compiler is used.
From rust-dev-guide

Reading this, I thought stage 0 would give the fastest compile time and errors I'm interested in, later on I can check stage1 and 2 before pushing.

And I tried without stage flag, got same error

[camelid]

And I tried without stage flag, got same error

Yeah, that part's probably because of #91927 (comment). Still, --stage 0 is incorrect here. Usually ./x.py build (with no arguments) will do the right thing.