Fix: handle partial write in stdio forwarding

[Feng Wang (chemwolf6922)]

Summary of the Pull Request

Partial writes on the WSL stdio relay were ignored. This could result in loss of data.
This PR adds the partial write handling logic.

PR Checklist

• Closes: [WSL 2.9.3] Large bracketed pastes silently lose 512-byte chunks at 4 KiB boundaries #41063
• Communication: I've discussed this with core contributors already. If work hasn't been agreed, this work might be rejected
• Tests: Added/updated if needed and all pass
• Localization: All end user facing strings can be localized
• Dev docs: Added/updated if needed
• Documentation updated: If checked, please file a pull request on our docs repo and link it here: #xxx

Detailed Description of the Pull Request / Additional comments

Validation Steps Performed

Tested manually by copying large amount of data to vim.

Regression tests are not added because the failure is hard to repro in e2e tests. Open to suggestions on the tests.

[Copilot]

Pull request overview

This PR fixes silent data loss in the Linux-side stdio relay by correctly handling short (partial) write() results when forwarding stdin into the child process. This aligns with the reported symptoms in #41063 where large pastes could drop data at 4 KiB boundaries due to ignored partial writes.

Changes:

• Detect partial writes to the child stdin FD and buffer the unwritten remainder in PendingStdin.
• Reuse the existing deferred-stdin write path to retry remaining bytes without blocking the relay loop.

[Copilot]

Pull request overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated no new comments.

[Ben Hillis (benhillis)]

This change looks correct to me, is it possible to have a unit test / stress test that validates the fix?

[Ben Hillis (benhillis)]

I think this will be difficult to write a unit test for, but the fix looks correct.