cp: support symlinks and timestamps on WASI

[DePasqualeOrg]

This is ready for review.

Two existing issues were found while preparing this PR. I'm flagging them for future work, since I didn't address them here:

• This PR fixes recursive timestamp preservation copying a traversal-updated directory atime. The current implementation causes about a 9% slowdown in affected benchmarks; avoiding it would require a different traversal design.
• On Android, cp -aL can suppress a hard-link failure for a dereferenced symlink. Two existing tests can still pass when the destination is missing because they only assert that it is not a symlink.

[xtqqczze]

Worth noting that we probably want to remove the filetime crate (#13808). If that would simplify these changes in any way, it might be worth considering here.

[oech3]

Suggested change

	// spell-checker:ignore (ToDO) copydir ficlone fiemap filestat ftruncate linkgs lstat nlink nlinks pathbuf pwrite reflink strs utimensat xattrs symlinked deduplicated advcpmv nushell IRWXG IRWXO IRWXU IRWXUGO IRWXU IRWXG IRWXO IRWXUGO sflag
	// spell-checker:ignore (ToDO) copydir fiemap filestat ftruncate linkgs lstat nlink nlinks pathbuf pwrite reflink strs utimensat xattrs symlinked deduplicated advcpmv nushell IRWXG IRWXO IRWXU IRWXUGO IRWXU IRWXG IRWXO IRWXUGO sflag

part of global config

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/cp/sparse-to-pipe. tests/cp/sparse-to-pipe is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/cut/bounded-memory (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/tail/symlink (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/misc/write-errors was skipped on 'main' but is now failing.

[codspeed-hq]

Merging this PR will degrade performance by 9.42%

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

❌ 2 regressed benchmarks
✅ 345 untouched benchmarks
⏩ 46 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	cp_preserve_metadata[(5, 4, 10)]	162.4 ms	179.9 ms	-9.76%
❌	Simulation	cp_archive_balanced_tree[(5, 4, 10)]	174.9 ms	192.4 ms	-9.08%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing DePasqualeOrg:codex/wasi-cp-symlinks-timestamps (c80eae6) with main (f2c5fec)}

Footnotes

1. 46 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩