Fixes rust-lang#111272.
With rust-lang#127912 merged, we now have all the infrastructure in place to support stack overflow detection in TLS destructors. This was not possible before because the signal stack was freed in the thread main function, thus a SIGSEGV afterwards would immediately crash. And on platforms without native TLS, the guard page address was stored in an allocation freed in a TLS destructor, so would not be available. rust-lang#127912 introduced the `local_pointer` macro which allows storing a pointer-sized TLS variable without allocation and the `thread_cleanup` runtime function which is called after all other code managed by the Rust runtime. This PR simply moves the signal stack cleanup to the end of `thread_cleanup` and uses `local_pointer` to store every necessary variable. And so, everything run under the Rust runtime is now properly protected against stack overflows.

...indows

The specific signal sent by the handler is unreliable when the stack overflow occurs in a TLS destructor. So just don't check the signal code, the only relevant part is that the program doesn't return a success.

...:open_native`

...kely to collide with symbol name mangling

As the comment at the top says, this file is not supposed to contain any
code. But some has crept in. This commit moves it out.

Fixed issue with usage of generics and moved feature gate to crate root
Removed const tag
Fixed alphabetical ordering of feature gate, added same to doctest
Removed crate-level declaration of feature gate control_flow_into_value
Used const_precise_live_drops to constify into_value without issue of a drop

..., r=Amanieu
std: detect stack overflows in TLS destructors on UNIX
Fixes rust-lang#111272.
With rust-lang#127912 merged, we now have all the infrastructure in place to support stack overflow detection in TLS destructors. This was not possible before because the signal stack was freed in the thread main function, thus a SIGSEGV afterwards would immediately crash. And on platforms without native TLS, the guard page address was stored in an allocation freed in a TLS destructor, so would not be available. rust-lang#127912 introduced the `local_pointer` macro which allows storing a pointer-sized TLS variable without allocation and the `thread_cleanup` runtime function which is called after all other code managed by the Rust runtime. This PR simply moves the signal stack cleanup to the end of `thread_cleanup` and uses `local_pointer` to store every necessary variable. And so, everything run under the Rust runtime is now properly protected against stack overflows.

Allow `IndexSlice` to be indexed by ranges.
This comes with some annoyances as the index type can no longer inferred from indexing expressions. The biggest offender for this is `IndexVec::from_fn_n(|idx| ..., n)` where the index type won't be inferred from the call site or any index expressions inside the closure.
My main use case for this is mapping a `Place` to `Range<Idx>` for value tracking where the range represents all the values the place contains.

...triddle
[rustdoc] Add new setting to wrap source code lines when too long
Fixes rust-lang#127334.
Wrapped lines look like this:
![image](https://github.com/user-attachments/assets/92006a27-ed1e-4beb-91f2-f453b72c5e1a)
It works in both source code pages and doc pages.
You can test it [here](https://rustdoc.crud.net/imperio/code-wrapping/bar/index.html).
r? `@notriddle`

...lize, r=Amanieu
Stabilize `unbounded_shifts`
This stabilizes and const-stabilizes `<iN>::unbounded_shl` and `<uN>::unbounded_shr` from rust-lang#129375.

...hrisDenton
Windows: use existing wrappers in `File::open_native`
Just a small improvement I've noticed - prevents accidents regarding `SetFileInformationByHandle` parameters.
Probably `@ChrisDenton` since we talked about it on discord :)

...Noratrieb
Fix documentation for unstable sort on slice
Fixes rust-lang#136665 

...hlin
Tighten `str-to-string-128690.rs``CHECK{,-NOT}`s to make it less likely to incorrectly fail with symbol name mangling
The `invoke` to match on to `CHECK` or `CHECK-NOT` (latest master) looks like
```llvm
 %_0.i.i.i.i.i.i.i.i.i.i.i.i.i1.i = invoke noundef zeroext i1 `@"_ZN42_$LT$str$u20$as$u20$core..fmt..Display$GT3ドルfmt17ha18033e7fb4f14fcE"(ptr` noalias noundef nonnull readonly align 1 %_3.val.i.i.i.i.i.i.i.i.i.i.i.i.i, i64 noundef %_3.val1.i.i.i.i.i.i.i.i.i.i.i.i.i, ptr noalias noundef nonnull align 8 dereferenceable(64) %formatter.i)
 to label %bb1.i unwind label %cleanup.i, !noalias !80
```
in the local `.ll` output.
This test incorrectly failed in rust-lang#137483 (comment) due to
```
// CHECK-NOT: {{(call|invoke).*}}fmt
```
matching against the unrelated call
```llvm
tail call void `@_RNvNtCseLfmtnDCoTB_5alloc7raw_vec12handle_error`
```
It's not pretty by any means, but...
r? `@saethlin`

...ble-control-flow-into-value, r=jhpratt
Added into_value function to ControlFlow<T, T>

..., r=compiler-errors
Move `impl` blocks out of `rustc_middle/src/mir/syntax.rs`
Best reviewed one commit at a time.