Reduce variance in sort benchmarks by reusing temp file

[naoNao89]

Bug existed since Sept 2024 but variance was below CodSpeed's threshold (~1-3%). On Oct 11, CI changed to run 19 benchmarks in parallel (f19d7ec, 6df73f0) instead of sequentially. This amplified /tmp contention, pushing variance to 2.64%—above detection threshold. The parallel execution exposed what was always there.

Solution

Move NamedTempFile::new() outside loop, matching sort_bench.rs pattern:

Eliminates filesystem noise while preserving correct semantics (sort overwrites file each iteration, no caching)

Notes
Affected PRs: #8935, #8936, #8927, #8926, #8930, #8932.

[codspeed-hq]

CodSpeed Performance Report

Merging #8939 will degrade performances by 3.36%

_{Comparing naoNao89:fix/sort-benchmark-variance (607383a) with main (0258583)}

Summary

❌ 2 regressions
✅ 104 untouched
⏩ 73 skipped¹

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

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
❌	du_balanced_tree[(5, 4, 10)]	9.1 ms	9.3 ms	-2.01%
❌	du_human_balanced_tree[(5, 4, 10)]	10.1 ms	10.5 ms	-3.36%

Footnotes

1. 73 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/misc/tee (fails in this run but passes in the 'main' branch)