tests(tail): fix race in untailable-symlink test

[eduardomourar]

Fixes intermittent CI failures in test_tail::test_follow_name_replaced_by_symlink_is_untailable, e.g. this run (https://github.com/uutils/coreutils/actions/runs/30700365364/job/91373194296?pr=13688), which failed with:

tail: 'watched' has become inaccessible: No such file or directory instead of the expected has been replaced with an untailable symbolic link.

Root cause

The test swapped the watched file for a symlink via remove() followed by symlink_file() — two separate syscalls.
Under polling (--use-polling, 0.1s interval), the poller could observe the transient "file missing" state in between,
reporting "has become inaccessible" instead of the intended "untailable symlink" message. The test also only checked the process was alive 500ms after spawn, without confirming tail's initial read had completed — under scheduler contention the symlink swap could happen before tail's first read of the file, which legitimately follows symlinks (no different from any other file open) since the untailable-symlink guard only applies to the follow loop, not the initial read.

Fix

• Wait for tail's initial content to actually appear in stdout before performing the swap.
• Perform the swap as an atomic rename() over a freshly created symlink, so the intermediate missing-file state is never observable to the poller.

[codspeed-hq]

Merging this PR will improve performance by 3.03%

⚠️ 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

⚡ 1 improved benchmark
✅ 344 untouched benchmarks
⏩ 46 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	cksum_crc32b	40.9 ms	39.7 ms	+3.03%

Tip

Curious why this is faster? Comment @codspeedbot explain why this is faster on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing eduardomourar:fix/tail-symlink-race-test (0079d4c) with main (0c8a3c7)}

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. ↩

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/tail/retry (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)
Skipping an intermittent issue tests/cut/bounded-memory (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/printf/printf-surprise is now being skipped but was previously passing.