Switch to bootstrapping from 1.27

It's possible the Float trait could be removed from core, but I couldn't tell whether it was intended to be removed or not. @SimonSapin may be able to comment more here; we can presumably also do that in a follow up PR as this one is already quite large.

In #49289, rustc was changed to emit metadata for binaries, which made
it so that the librustc.rmeta file created when compiling librustc was
overwritten by the rustc-main compilation. This commit renames the
rustc-main binary to avoid this problem.

https://github.com/rust-lang/cargo/issues/5524 has also been filed to
see if Cargo can learn to warn on this situation instead of leaving it
for the user to debug.

Rename trans to codegen everywhere.

Part of #45274.

 compiletest: Run revisions as independent tests.

Fixes #47604.

- The output of each test is now in its own directory.
- "auxiliary" output is now under the respective test directory.
- `stage_id` removed from filenames, and instead placed in the stamp file as a hash.  This helps keep path lengths down for Windows.

In brief, the new layout looks like this:
```
<build_base>/<relative_dir>/<testname>.<revision>.<mode>/
    stamp
    <testname>.err
    <testname>.out
    a (binary)
    auxiliary/lib<auxname>.dylib
    auxiliary/<auxname>/<auxname>.err
    auxiliary/<auxname>/<auxname>.out
```
(revision and mode are optional)

The aux dir, which previously had the `stage_id` embedded in it, was picking up remnants from previous runs.

- The output of each test is now in its own directory.
- "auxiliary" output is now under the respective test directory.
- `stage_id` removed from filenames, and instead placed in the stamp file as a hash.  This helps keep path lengths down for Windows.

In brief, the new layout looks like this:
```
<build_base>/<relative_dir>/<testname>.<revision>.<mode>/
    stamp
    <testname>.err
    <testname>.out
    a (binary)
    auxiliary/lib<auxname>.dylib
    auxiliary/<auxname>/<auxname>.err
    auxiliary/<auxname>/<auxname>.out
```
(revision and mode are optional)

rustc: Fix `crate` lint for single-item paths

This commit fixes recommending the `crate` prefix when migrating to 2018 for
paths that look like `use foo;` or `use {bar, baz}`

Closes #50660

Rollup of 17 pull requests

Successful merges:

 - #50170 (Implement From for more types on Cow)
 - #50638 (Don't unconditionally set CLOEXEC twice on every fd we open on Linux)
 - #50656 (Fix `fn main() -> impl Trait` for non-`Termination` trait)
 - #50669 (rustdoc: deprecate `#![doc(passes, plugins, no_default_passes)]`)
 - #50726 (read2: Use inner function instead of closure)
 - #50728 (Fix rustdoc panic with `impl Trait` in type parameters)
 - #50736 (env: remove unwrap in examples in favor of try op)
 - #50740 (Remove LazyBTreeMap.)
 - #50752 (Add missing error codes in libsyntax-ext asm)
 - #50779 (Make mutable_noalias and arg_align_attributes be tracked)
 - #50787 (Fix run-make wasm tests)
 - #50788 (Fix an ICE when casting a nonexistent const)
 - #50789 (Ensure libraries built in stage0 have unique metadata)
 - #50793 (tidy: Add a check for empty UI test files)
 - #50797 (fix a typo in signed-integer::from_str_radix())
 - #50808 (Stabilize num::NonZeroU*)
 - #50809 (GitHub: Stop treating Cargo.lock as a generated file.)

Failed merges:

Update Cargo

Unblocking PRs:
- rust-lang/cargo#5535 - Ignore tab in libtest output. (unblocks #50387)
- rust-lang/cargo#5537 - Remove -Zno-trans test. (unblocks #50615)
- rust-lang/cargo#5540 - Fix tests when CARGO_TARGET_DIR is set. (unblocks self)

Regression fixes:
- rust-lang/cargo#5503 - cargo rustc broken for tests in project with bins
- rust-lang/cargo#5520 - shared proc-macro dependency built incorrectly

Changes:
- rust-lang/cargo#5527 - Point Source Replacement to the Overriding Dependencies section
- rust-lang/cargo#5533 - Detail how to run locally-built nightly cargo
- rust-lang/cargo#5522 - Add option to set user-agent
- rust-lang/cargo#5519 - NFC: fix a couple of typos, found by codespell.
- rust-lang/cargo#5513 - Fix `panic` for binaries built during tests.
- rust-lang/cargo#5512 - simplify build_requirements
- rust-lang/cargo#5301 - Add --build-plan for 'cargo build'
- rust-lang/cargo#5460 - Be more conservative about which files are linked to the output dir.
- rust-lang/cargo#5509 - Use the new stable
- rust-lang/cargo#5507 - Does not print seconds fraction with minutes
- rust-lang/cargo#5498 - Bump to 0.29.0
- rust-lang/cargo#5497 - Mention +nightly in ARCHITECTURE.md

The PR fixes #50640.

GitHub: Stop treating Cargo.lock as a generated file.

We do want to inspect the changes to Cargo.lock, hiding the diff by default would make it easier to miss important details like https://github.com/rust-lang/rust/pull/50629#discussion_r187556602 and https://github.com/rust-lang/rust/pull/50696#pullrequestreview-119648156

Implement From for more types on Cow

This is basically https://github.com/rust-lang/rust/pull/48191, except that it should be implemented in a way that doesn't break third party crates.

Stabilize num::NonZeroU*

Tracking issue: https://github.com/rust-lang/rust/issues/49137

tidy: Add a check for empty UI test files

Check for empty `.stderr` and `.stdout` files in UI test directories.
Empty files could  still pass testing for `compile-pass` tests with no output
so they can get into the repo accidentally, but they are not necessary and can
be removed.

This is very much an in progress pull request. I'm having an issue with rustfmt. It wanted to reformat the entire file for almost every file by default. And when I run tidy it just errors out because it catches the empty files that are already in the repo.

My next step is goin got be to remove those empty file and see if running tidy again will actually reformat things outside of the context of `cargo fmt`

Fixes https://github.com/rust-lang/rust/issues/50785

fix a typo in signed-integer::from_str_radix()

a minor typo in docs.

Ensure libraries built in stage0 have unique metadata

Issue #50786 shows a case with local rebuild where the libraries built
by stage0 had the same suffix as stage0's own, and were accidentally
loaded by that stage0 rustc when compiling `librustc_trans`.

Now we set `__CARGO_DEFAULT_LIB_METADATA` to "bootstrap" during stage0,
rather than the release channel like usual, so the library suffix will
always be completely distinct from the stage0 compiler.

Fix an ICE when casting a nonexistent const

Fixes https://github.com/rust-lang/rust/issues/50599.

Fix run-make wasm tests

Fixes #50711

Make mutable_noalias and arg_align_attributes be tracked

Add missing error codes in libsyntax-ext asm

Remove LazyBTreeMap.

It was introduced in #50240 to avoid an allocation when creating a new
BTreeMap, which gave some speed-ups. But then #50352 made that the
default behaviour for BTreeMap, so LazyBTreeMap is no longer necessary.

env: remove unwrap in examples in favor of try op

Fix rustdoc panic with `impl Trait` in type parameters

Fixes #50702.

I'm not sure `impl Trait`s neither in arguments nor in return types are supposed to work, though.

read2: Use inner function instead of closure

Very minor thing, but there doesn't appear to be a reason to use a closure here.

Generated code is identical in my tests, but I believe it's clearer that nothing from the environment is being used.

Fix conversion from Miri Value to ConstValue

This fixes an error compiling the `immeta` 0.3.6 crate. https://github.com/rust-lang/rust/issues/50707 may be fixed too.

r? @oli-obk

Check for empty `.stderr` and `.stdout` files in UI test directories.
Empty files could  still pass testing for `compile-pass` tests with no output
so they can get into the repo accidentally, but they are not necessary and can
be removed.

Reenable the MergeFunctions pass

The crash that happened in #23566 doesn't happen anymore with the LLVM mergefunc
pass enabled and it hugely reduces code size (for example it shaves off 10% of the
final Servo executable). This patch reenables it.

For those wondering, [here are the docs from LLVM about this pass](http://llvm.org/docs/MergeFunctions.html).

Tracking issue: https://github.com/rust-lang/rust/issues/49137

It is now an implementation detail of ptr::NonNull and num::NonZero*

We do want to inspect the changes to Cargo.lock, hiding the diff by default
would miss important details like
https://github.com/rust-lang/rust/pull/50629#discussion_r187556602