sort: propagate temporary file write errors

[tasodoufu]

Summary

Fixes #14377.

Reproduction

When external sort spills a chunk to a temporary file, a write failure (for example, ENOSPC/EFBIG) reached write_lines, whose write_all calls unconditionally used unwrap(). The process therefore panicked instead of reporting the I/O error and exiting non-zero.

Cause

The temporary-file chunk writer discarded the io::Result from both line and separator writes.

Fix

Propagate write errors from write_lines to the external-sort path with ?, preserving the existing error handling and cleanup flow. Added a regression unit test using a writer that returns StorageFull.

Validation

• cargo fmt --check
• cargo test -p uu_sort
• cargo clippy -p uu_sort --all-targets -- -D warnings
• git diff --check

The focused test and full uu_sort unit-test suite pass (29 tests).

Not verified

A real full/size-limited filesystem was not mounted in this environment; the regression test deterministically exercises the same failing Write path.

AI assistance was used to help inspect the issue and prepare this patch; the submitted change was reviewed and verified against the repository's contributing and AI-policy requirements.

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/date/date-locale-hour (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/tail/pid-pipe (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/tail/symlink (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/tail/tail-n0f (fails in this run but passes in the 'main' branch)
Note: The gnu test tests/seq/seq-epipe is now being skipped but was previously passing.
Note: The gnu test tests/tail/pipe-f is now being skipped but was previously passing.
Congrats! The gnu test tests/rm/many-dir-entries-vs-OOM is now passing!