mktemp: Fix template validation to require trailing consecutive Xs

[Xylphy]

Closes #10187

The validation logic now checks only the trailing placeholder block, as GNU mktemp requires. It walks backward from the end of the template and counts consecutive X bytes. The template is accepted only if the trailing run has length ≥ 3.

Problem
The previous implementation accepted templates containing "XXX" anywhere, which incorrectly allowed cases where the template does not end with a sufficient run of Xs (e.g. XXX_foo) or where an earlier "XXX" masked an invalid suffix.

Implementation details

• Scans only the suffix: starts at s.len() and walks left while bytes are b'X'.
• No allocations; runtime proportional to the trailing X run.
• UTF-8 safe: we only compare ASCII X bytes and compute indices from the end boundary.

Result

• Rejects templates without ≥3 trailing Xs
• Aligns behavior with GNU mktemp / coreutils validation rules

[sylvestre]

Please add tests to make sure we don't regress in the future

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/fmt/non-space. tests/fmt/non-space is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/misc/mktemp. tests/misc/mktemp is passing on 'main'. Maybe you have to rebase?
GNU test failed: tests/sort/sort-debug-warn. tests/sort/sort-debug-warn is passing on 'main'. Maybe you have to rebase?

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/misc/mktemp. tests/misc/mktemp is passing on 'main'. Maybe you have to rebase?
Skipping an intermittent issue tests/tty/tty-eof (passes in this run but fails in the 'main' branch)

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/shuf/shuf-reservoir (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/sort/sort-stale-thread-mem (fails in this run but passes in the 'main' branch)

[codspeed-hq]

CodSpeed Performance Report

Merging this PR will degrade performance by 27.39%

_{Comparing Xylphy:mktemp_consecutive_x (b74bdec) with main (2c75e71)}

Summary

⚡ 2 improved benchmarks
❌ 7 regressed benchmarks
✅ 273 untouched benchmarks
⏩ 38 skipped benchmarks¹

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Memory	sort_unique_locale[500000]	33.6 MB	39.8 MB	-15.51%
❌	Memory	sort_accented_data[500000]	22.1 MB	28.3 MB	-21.79%
❌	Memory	sort_numeric[500000]	75.5 MB	79.2 MB	-4.67%
❌	Memory	sort_mixed_data[500000]	22.9 MB	27.1 MB	-15.71%
❌	Memory	sort_key_field[500000]	47.8 MB	51.8 MB	-7.62%
⚡	Memory	sort_ascii_utf8_locale	6.7 MB	6.2 MB	+7.93%
❌	Memory	sort_long_line[160000]	712.6 KB	981.4 KB	-27.39%
❌	Memory	sort_ascii_only[500000]	22.2 MB	28.3 MB	-21.77%
⚡	Memory	du_deep_tree[(100, 3)]	157.9 KB	146.6 KB	+7.72%

Footnotes

1. 38 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. ↩

[cakebaker]

Thanks for your PR!