Change RangeInclusive to a three-field struct.

Fix #45222.

This PR also reverts #48012 (i.e. removed the `try_fold`/`try_rfold` specialization for `RangeInclusive`) because LLVM no longer has trouble recognizing a RangeInclusive loop.

Use fast TLS on Fuchsia

I'm not sure why Fuchsia was separated here, but we provide these symbols, and tests are passing in QEMU with this change. cc @raphlinus.

r? @alexcrichton

rustc: Lint against `#[macro_use]` in 2018 idioms

This commit adds a lint to the compiler to warn against the `#[macro_use]`
directive as part of the `rust_2018_idioms` lint. This lint is turned off by
default and is only enabled when the `use_extern_macros` feature is also
enabled.

The lint here isn't fully fleshed out as it's just a simple warning rather than
suggestions of how to actually import the macro, but hopefully it's a good base
to start from!

cc #52043

When the index is not PartialOrd, always treat the range as empty.

Fix #45222.

Ensure StorageDead is created even if variable initialization fails

Rebase and slight cleanup of https://github.com/rust-lang/rust/pull/51109
Fixes https://github.com/rust-lang/rust/issues/49232

r? @EddyB

This commit adds a lint to the compiler to warn against the `#[macro_use]`
directive as part of the `rust_2018_idioms` lint. This lint is turned off by
default and is only enabled when the `use_extern_macros` feature is also
enabled.

The lint here isn't fully fleshed out as it's just a simple warning rather than
suggestions of how to actually import the macro, but hopefully it's a good base
to start from!

cc #52043

Add ExactChunks::remainder and ExactChunks::into_remainder

These allow to get the leftover items of the slice that are not being
iterated as part of the iterator due to not filling a complete chunk.

The mutable version consumes the slice because otherwise we would either
a) have to borrow the iterator instead of taking the lifetime of
the underlying slice, which is not what *any* of the other iterator
functions is doing, or
b) would allow returning multiple mutable references to the same data

The current behaviour of consuming the iterator is consistent with
IterMut::into_slice for the normal iterator.

----

This is related to https://github.com/rust-lang/rust/issues/47115#issuecomment-392685177 and the following comments.

While there the discussion was first about a way to get the "tail" of the iterator (everything from the slice that is still not iterated yet), this gives kind of unintuitive behaviour and is inconsistent with how the other slice iterators work.

Unintuitive because the `next_back` would have no effect on the tail (or otherwise the tail could not include the remainder items), inconsistent because a) generally the idea of the slice iterators seems to be to only ever return items that were not iterated yet (and don't provide a way to access the same item twice) and b) we would return a "flat" `&[T]` slice but the iterator's shape is `&[[T]]` instead, c) the mutable variant would have to borrow from the iterator instead of the underlying slice (all other iterator functions borrow from the underlying slice!)

As such, I've only implemented functions to get the remainder. This also allows the implementation to be completely safe still (and around slices instead of raw pointers), while getting the tail would either be inefficient or would have to be implemented around raw pointers.

CC @kerollmops

make pretty source comparison check be fatal (fixes #52255)

This is not ready for merging because it reveals (at least) two regressions in the pretty suite. Should I attempt to fix those in this PR also?

Rollup of 9 pull requests

Successful merges:

 - #51816 (bootstrap: write texts to a .tmp file first for atomicity)
 - #51912 (impl Clone for Box<CStr>, Box<OsStr>, Box<Path>)
 - #52164 (use proper footnote syntax for references)
 - #52220 (Deny bare trait objects in `src/bootstrap`)
 - #52276 (rustc: Verify #[proc_macro] is only a word)
 - #52277 (Uncapitalize "If")
 - #52287 (Deny bare trait objects in src/librustc_resolve)
 - #52295 (Deny bare trait objects in src/libsyntax_ext)
 - #52298 (make reference to dirs crate clickable in terminals)

Failed merges:

r? @ghost

make reference to dirs crate clickable in terminals

Currently I have to copy-paste the link; with this change I can just click it right in my terminal window.

Deny bare trait objects in src/libsyntax_ext

Enforce `#![deny(bare_trait_objects)]` in `src/libsyntax_ext`.

Deny bare trait objects in src/librustc_resolve

Enforce `#![deny(bare_trait_objects)]` in `src/librustc_resolve`.

Uncapitalize "If"

rustc: Verify #[proc_macro] is only a word

... and perform the same verification for #[proc_macro_attribute], currently
neither of these attributes take any arguments.

Closes #52273

Deny bare trait objects in `src/bootstrap`

Enforce `#![deny(bare_trait_objects)]` in `src/bootstrap`.

use proper footnote syntax for references

The previous syntax was causing rustdoc to interpret them as links.

impl Clone for Box<CStr>, Box<OsStr>, Box<Path>

Implements #51908.

bootstrap: write texts to a .tmp file first for atomicity

If you are using a hard-linked file as your config.toml, this change will affect the way other instances of the file is modified.
The original version would modify all other instances whereas the new version will leave others unchanged, reducing the ref count by one.

rustc: Search all derives for inert attributes

This commit fixes an apparent mistake in librustc_resolve where when the
`proc_macro` feature is enabled (or `rust_2018_preview`) the resolution of
custom attributes for custom derive was tweaked. Previously when an attribute
failed to resolve it was attempted to locate if there is a custom derive also in
scope which declares the attribute, but only the first custom derive directive
was search.

Instead this commit fixes the loop to search all custom derive invocations
looking for any which register the attribute in question.

Closes #52219

Patch clippy_lints to use the in-tree clippy as well.

Should fix the most recent RLS build break.

Remove rustdoc's plugins feature

This fixes CVE-2018-1000622.

https://cve.mitre.org/cgi-bin/cvename.cgi?name=%20CVE-2018-1000622

r? @QuietMisdreavus @GuillaumeGomez

rustc_codegen_llvm: replace the first argument early in FnType::new_vtable.

Fixes #51907 by removing the vtable pointer before the `ArgType` is even created.
This allows any ABI to support trait object method calls, regardless of how it passes `*dyn Trait`.

r? @nikomatsakis

... and perform the same verification for #[proc_macro_attribute], currently
neither of these attributes take any arguments.

Closes #52273

Inject clippy into the rls again

Also makes sure we actually point to the local rustfmt

r? @nrc

cc @Manishearth

Rollup of 14 pull requests

Successful merges:

 - #51614 (Correct suggestion for println)
 - #51952 ( hygiene: Decouple transparencies from expansion IDs)
 - #52193 (step_by: leave time of item skip unspecified)
 - #52207 (improve error message shown for unsafe operations)
 - #52223 (Deny bare trait objects in in src/liballoc)
 - #52224 (Deny bare trait objects in in src/libsyntax)
 - #52239 (Remove sync::Once::call_once 'static bound)
 - #52247 (Deny bare trait objects in in src/librustc)
 - #52248 (Deny bare trait objects in in src/librustc_allocator)
 - #52252 (Deny bare trait objects in in src/librustc_codegen_llvm)
 - #52253 (Deny bare trait objects in in src/librustc_data_structures)
 - #52254 (Deny bare trait objects in in src/librustc_metadata)
 - #52261 (Deny bare trait objects in in src/libpanic_unwind)
 - #52265 (Deny bare trait objects in in src/librustc_codegen_utils)

Failed merges:

r? @ghost

Deny bare trait objects in in src/librustc_codegen_utils

Enforce `#![deny(bare_trait_objects)]` in `src/librustc_codegen_utils`.

Deny bare trait objects in in src/libpanic_unwind

Enforce `#![deny(bare_trait_objects)]` in `src/libpanic_unwind`.

Deny bare trait objects in in src/librustc_metadata

Enforce `#![deny(bare_trait_objects)]` in `src/librustc_metadata`.