Implement `Read` and `Write` traits for `BorrowedFd` and `OwnedFd`

Proposal

Problem statement

When working with FFI functions, we often need to perform IO operations on file descriptors (FDs) produced by foreign language APIs. To reduce unsafe code usage, we often need to wrap FDs with BorrowedFd or OwnedFd. But currently they do not implement Read or Write traits, so if we want to perform IO operations, we have to convert them to types like File or UnixStream. If we only have a BorrowedFd object, we have to call BorrowedFd::try_clone_to_owned in order to complete the conversion, which might be suboptimal.

Solution sketch

Implement std::io::Read and std::io::Write traits for the following types:

• std::os::fd::BorrowedFd
• std::os::fd::OwnedFd
• &std::os::fd::BorrowedFd
• &std::os::fd::OwnedFd

Alternatives

• An alternative approach would be to use third-party crates like libc and nix for IO operations. However, we lose the ability to use standard library IO utilities such as std::io::{copy, BufReader, BufWriter}.
• Developers could implement their own wrapper types like struct BorrowedReadableFd<'a>(Borrowed<Fd<'a>>), but it could cause duplicated code if multiple crates have the same need.

What happens now?

This issue contains an API change proposal (or ACP) and is part of the libs-api team feature lifecycle. Once this issue is filed, the libs-api team will review open proposals as capability becomes available. Current response times do not have a clear estimate, but may be up to several months.

Possible responses

The libs team may respond in various different ways. First, the team will consider the problem (this doesn't require any concrete solution or alternatives to have been proposed):

• We think this problem seems worth solving, and the standard library might be the right place to solve it.
• We think that this probably doesn't belong in the standard library.

Second, if there's a concrete solution:

• We think this specific solution looks roughly right, approved, you or someone else should implement this. (Further review will still happen on the subsequent implementation PR.)
• We're not sure this is the right solution, and the alternatives or other materials don't give us enough information to be sure about that. Here are some questions we have that aren't answered, or rough ideas about alternatives we'd want to see discussed.

Start with the proposal and the linked feature lifecycle guidance; no implementation files or tests are identified. Follow the libs-api team’s review of whether the problem and proposed trait implementations belong in the standard library. Done requires an approved direction before implementation can be scoped.