perf(nl): optimize line numbering by using itoa and direct writing

[CrazyRoka]

This PR optimizes the nl command's performance by refactoring how line numbers are formatted and written. Previously, the code used the format! macro for every line, which involves heap allocations and runtime parsing of format strings.

By integrating the itoa crate and writing directly to the output BufWriter, we significantly reduce the overhead per line.

[github-actions]

GNU testsuite comparison:

Congrats! The gnu test tests/basenc/bounded-memory is now passing!

[codspeed-hq]

Merging this PR will improve performance by ×2.2

⚡ 2 improved benchmarks
✅ 282 untouched benchmarks
⏩ 38 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	nl_large_file[10]	58.8 ms	26.9 ms	×2.2
⚡	Simulation	nl_many_lines[100000]	46.1 ms	20.8 ms	×2.2

---

_{Comparing CrazyRoka:optimize-nl-formatting (bccfda0) with main (25a43c6)}

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. ↩

[ChrisDryden]

Some of the test failures were because of a broken main, rebasing it

[github-actions]

GNU testsuite comparison:

Skipping an intermittent issue tests/misc/usage_vs_getopt (passes in this run but fails in the 'main' branch)
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]

Can you used unsigned_abs? I think this can cause a panic

[ChrisDryden]

Actually this is unrelated to the original change and was already there lgtm

[ChrisDryden]

Thanks!