ci: enable memory profiling again

[not-matthias]

Hey! We've fixed an issue in memory profiling which caused benchmarks to have huge variance, and the results are very stable now.

I've run your benchmark CI 10 times on our fork, and there's now 0% variance across those runs: https://codspeed.io/AvalancheHQ/coreutils-1/benchmarks?q=mode%3Amemory

Let me know if you run into any other issues or if you have further questions!

[oech3]

Needs #10388

[github-actions]

GNU testsuite comparison:

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

[ChrisDryden]

Thanks! Very exited to get this used more often.

Needs #10388

I'm not sure that this is the case, I understand that theoretically this can impact the metrics but when we were testing it the impacts to the memory profiles were not measurable. As long as the memory profiles are consistent when we test I think it should be good to go.

[codspeed-hq]

Merging this PR will improve performance by ×370

⚡ 125 improved benchmarks
✅ 159 untouched benchmarks
⏩ 38 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Memory	sort_ascii_utf8_locale	5.6 MB	5.1 MB	+10.26%
⚡	Memory	sort_ascii_only[500000]	25.5 MB	21.6 MB	+18.25%
⚡	Memory	sort_numeric[500000]	48.7 MB	44.8 MB	+8.75%
⚡	Memory	sort_dictionary_order[500000]	29.7 MB	22 MB	+35.09%
⚡	Memory	sort_ascii_c_locale	6.1 MB	5.1 MB	+19.88%
⚡	Memory	sort_german_de_locale	3.6 MB	2.7 MB	+32.65%
⚡	Memory	sort_accented_data[500000]	25.5 MB	21.6 MB	+18.23%
⚡	Memory	sort_key_field[500000]	32.8 MB	28.9 MB	+13.57%
⚡	Memory	sort_unique_locale[500000]	36.9 MB	33 MB	+11.93%
⚡	Memory	sort_german_c_locale	3.6 MB	2.7 MB	+32.65%
⚡	Memory	sort_mixed_data[500000]	25.8 MB	21.9 MB	+17.96%
⚡	Memory	join_unicode_locale	57.2 KB	46.2 KB	+23.82%
⚡	Memory	join_partial_overlap	55.5 KB	46.2 KB	+20.19%
⚡	Memory	join_french_locale	55.6 KB	46.2 KB	+20.3%
⚡	Memory	join_full_match	55.6 KB	46.2 KB	+20.47%
⚡	Memory	join_custom_separator	96.9 KB	83.9 KB	+15.47%
⚡	Memory	numfmt_large_numbers_si[10000]	5 MB	4 MB	+23.7%
⚡	Memory	numfmt_round_modes[("down", 10000)]	4.9 MB	3.9 MB	+25.33%
⚡	Memory	numfmt_round_modes[("up", 10000)]	4.9 MB	3.9 MB	+25.34%
⚡	Memory	numfmt_to_iec[10000]	5 MB	3.9 MB	+27.01%
...	...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

---

_{Comparing AvalancheHQ:enable-memory-profiling (3d43765) with main (fb6a2bb)}

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. ↩

[RenjiSann]

Thank you !