Rollup of 8 pull requests

Successful merges:

 - #88205 (Add Explanation For Error E0772)
 - #92274 (Add `intrinsics::const_deallocate`)
 - #93236 (Make `NonNull::new` `const`)
 - #93299 (Fix dot separator when there is no source link)
 - #93410 (kmc-solid: Implement `net::FileDesc::duplicate`)
 - #93424 (fix nit)
 - #93431 (remove unused `jemallocator` crate)
 - #93453 (Add GUI theme change test)

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup

Add GUI theme change test

r? `@jsha`

remove unused `jemallocator` crate

When it was noticed that the rustc binary wasn't actually using jemalloc via `#[global_allocator]` and that was removed, the dependency remained.

Tests pass locally with a `jemalloc = true` build, but I'll trigger a try build to ensure I haven't missed an edge-case somewhere.

r? ```@ghost``` until that completes

fix nit

kmc-solid: Implement `net::FileDesc::duplicate`

This PR implements `std::sys::solid::net::FileDesc::duplicate`, which was accidentally left out when this target was added by #86191.

Fix dot separator when there is no source link

Fixes #92973.

We did well adding this test since there was a bug:

![Screenshot from 2022-01-25 17-05-48](https://user-images.githubusercontent.com/3050060/151016535-39b45f52-e1e0-4963-ad19-532e24ec4c9b.png)

r? `@jsha`

Make `NonNull::new` `const`

Tracking issue: #93235

Add `intrinsics::const_deallocate`

Tracking issue: #79597
Related: #91884

This allows deallocation of a memory allocated by `intrinsics::const_allocate`. At the moment, this can be only used to reduce memory usage, but in the future this may be useful to detect memory leaks (If an allocated memory remains after evaluation, raise an error...?).

Add Explanation For Error E0772

I've added an error explanation for the error code E0772.

Assists with #61137

Add tests for three old ICEs

Closes #84044
Closes #91594
Closes #89066

Bump libc and fix remove_dir_all on Fuchsia after CVE fix

With the previous `is_dir` impl, we would attempt to unlink
a directory in the None branch, but Fuchsia supports returning
ENOTEMPTY from unlinkat() without the AT_REMOVEDIR flag because
we don't currently differentiate unlinking files and directories
by default.

On the Fuchsia side I've opened https://fxbug.dev/92273 to discuss
whether this is the correct behavior, but it doesn't seem like
addressing the error code is necessary to make our tests happy.

Depends on https://github.com/rust-lang/libc/pull/2654 since we
apparently haven't needed to reference DT_UNKNOWN before this.

Move tier-2 (without host tools) apple targets to separate builder

One-off (likely fairly unreliable, but give some idea) measurements:

* dist-apple-various (new):  2h10m
* dist-x86_64-apple: 2h55m -> 2h36m (cutting roughly 20 minutes)

The ICE from #84044 no longer occurs.

Rollup of 10 pull requests

Successful merges:

 - #92611 (Add links to the reference and rust by example for asm! docs and lints)
 - #93158 (wasi: implement `sock_accept` and enable networking)
 - #93239 (Add os::unix::net::SocketAddr::from_path)
 - #93261 (Some unwinding related cg_ssa cleanups)
 - #93295 (Avoid double panics when using `TempDir` in tests)
 - #93353 (Unimpl {Add,Sub,Mul,Div,Rem,BitXor,BitOr,BitAnd}<$t> for Saturating<$t>)
 - #93356 (Edit docs introduction for `std::cmp::PartialOrd`)
 - #93375 (fix typo `documenation`)
 - #93399 (rustbuild: Fix compiletest warning when building outside of root.)
 - #93404 (Fix a typo from #92899)

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup

With the previous `is_dir` impl, we would attempt to unlink
a directory in the None branch, but Fuchsia supports returning
ENOTEMPTY from unlinkat() without the AT_REMOVEDIR flag because
we don't currently differentiate unlinking files and directories
by default.

On the Fuchsia side I've opened https://fxbug.dev/92273 to discuss
whether this is the correct behavior, but it doesn't seem like
addressing the error code is necessary to make our tests happy.

Updates std's libc crate to include DT_UNKNOWN for Fuchsia.

backport llvm fix for issue 91671.

fix #91671

Fix a typo from #92899

Just happened to notice this in passing

rustbuild: Fix compiletest warning when building outside of root.

This fixes a warning that would happen when passing arguments to compiletest (like `x.py test src/test/ui`) when running `x.py` outside of the root source directory. For example, the CI builders do this, which causes a confusing warning message. This also fixes it so that passing a full path works (like `x.py test src/test/ui/hello.rs`) in the same scenario (previously it would just ignore the `hello.rs` part).

fix typo `documenation`

Edit docs introduction for `std::cmp::PartialOrd`

This makes `PartialOrd` consistent with the other three traits in this module, which all include links to their corresponding mathematical concepts on Wikipedia.

<img width="500" alt="Screen Shot 2022-01-26 at 10 24 23 PM" src="https://user-images.githubusercontent.com/19642016/151291720-decd85ed-cd6e-4be0-84a9-619b98ceb386.png">

Unimpl {Add,Sub,Mul,Div,Rem,BitXor,BitOr,BitAnd}<$t> for Saturating<$t>

Tracking issue #92354

Analog to 9648b313 #93208 reduce `saturating_int_assign_impl` (#93208) to:

```rust
let mut value = Saturating(2u8);
value += 3u8;
value -= 1u8;
value *= 2u8;
value /= 2u8;
value %= 2u8;
value ^= 255u8;
value |= 123u8;
value &= 2u8;
```

See https://github.com/rust-lang/rust/pull/93208#issuecomment-1022564429

Avoid double panics when using `TempDir` in tests

`TempDir` could panic on drop if `remove_dir_all` returns an error. If this happens while already panicking, the test process would abort and therefore not show the test results.

This PR tries to avoid such double panics.

Some unwinding related cg_ssa cleanups

These should make it a bit easier for alternative codegen backends to implement unwinding.

Add os::unix::net::SocketAddr::from_path

Creates a new SocketAddr from a path, supports both regular paths and
abstract namespaces.

Note that `SocketAddr::from_abstract_namespace` could be removed after this as `SocketAddr::unix` also supports abstract namespaces.

Updates #65275
Unblocks https://github.com/tokio-rs/mio/issues/1527

r? `@m-ou-se`

wasi: implement `sock_accept` and enable networking

With the addition of `sock_accept()` to snapshot1, simple networking via a passed `TcpListener` is possible. This PR implements the basics to make a simple server work.

See also:
* [wasmtime tracking issue](https://github.com/bytecodealliance/wasmtime/issues/3730)
* [wasmtime PR](https://github.com/bytecodealliance/wasmtime/pull/3711)

TODO:
* [ ] Discussion of `SocketAddr` return value for `::accept()`

```rust
        Ok((
            TcpStream::from_inner(unsafe { Socket::from_raw_fd(fd as _) }),
            // WASI has no concept of SocketAddr yet
            // return an unspecified IPv4Addr
            SocketAddr::new(Ipv4Addr::UNSPECIFIED.into(), 0),
        ))
```

Add links to the reference and rust by example for asm! docs and lints

These were previously removed in #91728 due to broken links.

cc ``@ehuss`` since this updates the rust-by-example submodule

Fix debuginfo for pointers/references to unsized types

This PR makes the compiler emit fat pointer debuginfo in all cases. Before, we sometimes got thin-pointer debuginfo, making it impossible to fully interpret the pointed to memory in debuggers. The code is actually cleaner now, especially around generation of trait object pointer debuginfo.

Fixes https://github.com/rust-lang/rust/issues/92718

~~Blocked on https://github.com/rust-lang/rust/pull/92729.~~

With the addition of `sock_accept()` to snapshot1, simple networking via
a passed `TcpListener` is possible. This patch implements the basics to
make a simple server work.
Signed-off-by: NHarald Hoyer <harald@profian.com>