Co-authored-by: Konrad Borowski <konrad@borowski.pw>

Renames File::with_options to File::options, per consensus in
rust-lang#65439, and stabilizes it.

Rename WASI's `FileTypeExt::is_character_device` to
`FileTypeExt::is_char_device`, for consistency with the Unix
`FileTypeExt::is_char_device`.

Also, add a `FileTypeExt::is_socket` function, for consistency with the
Unix `FileTypeExt::is_socket` function.

These are working well for *unsigned* types, for the the signed ones there are a bunch of questions about what the semantics and API should be.  And for the main "helpers for big integer implementations" use, there's no need for the signed versions anyway.

And there are plenty of other methods which exist for unsigned types but not signed ones, like `next_power_of_two`, so this isn't unusual.

Fixes 90541

Tweak the `options!` macro to allow for -Z and -C options with the same
name without generating conflicting internal parsing functions.

Split out of the commit stabilizing -Z strip as -C strip.

Leave -Z strip available temporarily as an alias, to avoid breaking
cargo until cargo transitions to using -C strip. (If the user passes
both, the -C version wins.)

Stabilize File::options()

Renames File::with_options to File::options, per consensus in
rust-lang#65439, and stabilizes it.

…llible, r=yaahc

Implement `Termination` for `Result<Infallible, E>`

As noted in rust-lang#43301, `Result<!, E>` is not usable on stable.

…leywiser

Stabilize -Z strip as -C strip

Leave -Z strip available temporarily as an alias, to avoid breaking
cargo until cargo transitions to using -C strip.

…-Simulacrum

Fix standard library test with read_link

closes rust-lang#90669
resolve this issue by comparing between Paths instead of strs

Android is not GNU

For a long time, the Android targets had `target_env=""`, but this changed to `"gnu"` in Rust 1.49.0. I tracked this down to rust-lang#77729 which started setting `"gnu"` in the `linux_base` target options, and this was inherited by `android_base`. Then rust-lang#78929 split the env into `linux_gnu_base`, but `android_base` was also changed to follow that. Android was not specifically mentioned in either pull request, so I believe this was an accident. Moving it back to `linux_base` will use an empty `env` again.

r? ````@Mark-Simulacrum````
cc ````@petrochenkov````

…evice, r=alexcrichton

Rename WASI's `is_character_device` to `is_char_device`.

Rename WASI's `FileTypeExt::is_character_device` to
`FileTypeExt::is_char_device`, for consistency with the Unix
`FileTypeExt::is_char_device`.

Also, add a `FileTypeExt::is_socket` function, for consistency with the
Unix `FileTypeExt::is_socket` function.

r? `@alexcrichton`

Move some tests to more reasonable directories - 9

cc rust-lang#73494
r? `@petrochenkov`

…rs, r=joshtriplett

Remove bigint_helper_methods for *signed* types

This PR inspired by `@cuviper's` comment @ rust-lang#90541 (comment)

These are working well for *unsigned* types, so keep those, but for the the *signed* ones there are a bunch of questions about what the semantics and API should be.  For the main "helpers for big integer implementations" use, there's no need for the signed versions anyway.  There are plenty of other methods which exist for unsigned types but not signed ones, like `next_power_of_two`, so this isn't unusual.

Fixes rust-lang#90541
Tracking issue rust-lang#85532

…li-obk

fix ICE on Miri/CTFE copy of half a pointer

Fixes rust-lang/miri#1910
r? `````@oli-obk`````

…orkingjubilee

disable portable SIMD tests in Miri

Until rust-lang/miri#1912 is resolved, we'll have to skip these tests in Miri.