ls: replace recursion with depth-first search

[Alonely0]

Changes ls to use a Depth-First Search (DFS) algorithm instead of recursion. Reopened from #11384 because I closed it by mistake.

Fixes #8725 and should help towards fixing #11215; this also opens the door for greater optimizations that fully fix the latter.

[codspeed-hq]

Merging this PR will degrade performance by 92.89%

⚠️ 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

⚡ 7 improved benchmarks
❌ 5 regressed benchmarks
✅ 286 untouched benchmarks
⏩ 48 skipped benchmarks¹

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

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Memory	ls_recursive_deep_tree[(200, 2)]	570.5 KB	143.2 KB	×4
❌	Memory	ls_recursive_balanced_tree[(6, 4, 15)]	129.2 KB	1,817.9 KB	-92.89%
❌	Memory	ls_recursive_long_all_balanced_tree[(6, 4, 15)]	310.5 KB	1,986.5 KB	-84.37%
⚡	Memory	ls_recursive_long_all_deep_tree[(100, 4)]	448.1 KB	126 KB	×3.6
⚡	Memory	ls_recursive_long_all_wide_tree[(15000, 1500)]	14.1 MB	9.9 MB	+42.58%
❌	Memory	ls_recursive_long_all_mixed_tree	115.9 KB	132.4 KB	-12.51%
⚡	Memory	ls_recursive_wide_tree[(10000, 1000)]	8 MB	6.6 MB	+21.09%
❌	Memory	ls_recursive_mixed_tree	115.3 KB	133.1 KB	-13.33%
⚡	Simulation	ls_recursive_deep_tree[(200, 2)]	1.8 ms	1.7 ms	+3.55%
❌	Simulation	ls_recursive_balanced_tree[(6, 4, 15)]	52.8 ms	57 ms	-7.32%
⚡	Simulation	ls_recursive_long_all_deep_tree[(100, 4)]	3.1 ms	2.9 ms	+6.19%
⚡	Simulation	ls_recursive_long_all_wide_tree[(15000, 1500)]	143.1 ms	136 ms	+5.21%

---

_{Comparing Alonely0:ls-dfs-fix (c913b43) with main (3b3d5a7)²}

Footnotes

1. 48 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. ↩

2. No successful run was found on main (0474e55) during the generation of this report, so 3b3d5a7 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/pr/bounded-memory (fails in this run but passes in the 'main' branch)
Congrats! The gnu test tests/tail/retry is no longer failing!
Note: The gnu test tests/env/env-signal-handler was skipped on 'main' but is now failing.

[github-actions]

GNU testsuite comparison:

Note: The gnu test tests/misc/write-errors was skipped on 'main' but is now failing.

[Alonely0]

@sylvestre I've done some modifications to keep memory usage stable across all benchmarks, except those with combinatorial explosion (which are not very realistic, but worth taking a look at; I have already pushed an optimization for them). I'll work on perf and maybe do some refactoring in another PR, if that's good by you. I think this is a good baseline for future work.