tsort: handle stdout write errors without panicking

[dragutreis]

Fixes #12006

Avoid a panic in tsort when stdout is unwritable (for example, redirected to /dev/full). Previously an unchecked println! was used which could panic on write errors. Replaced that with a writeln! wrapped in error handling so the program continues generating cycle diagnostics even if writing to stdout fails.

Cycle diagnostics are still always written to stderr; any write errors are recorded and cause the program to exit with status 1 after processing completes.

Tested with /etc/pacman.conf and printf "a b\nb a\n" | tsort > /dev/full in both cases.

[github-actions]

GNU testsuite comparison:

Skipping an intermittent issue tests/tail/tail-n0f (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/cut/bounded-memory is now passing!
Congrats! The gnu test tests/printf/printf-surprise is now passing!

[codspeed-hq]

Merging this PR will not alter performance

✅ 97 untouched benchmarks
⏩ 261 skipped benchmarks¹

---

_{Comparing dragutreis:fix/tsort-stdout-write-error (cf31a00) with main (ead174f)²}

Footnotes

1. 261 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 (5316f58) during the generation of this report, so ead174f was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[xtqqczze]

I guess it would be better if run_tsort returned Result<(), TsortError> instead

Suggested change

	g.run_tsort().map_err(|e| USimpleError::new(1, format!("write error: {e}")))?;
	g.run_tsort()?;

You'd need to add a TsortError variant:

    #[error("{}", translate!("tsort-error-write", "error" => strip_errno(.0)))]
    Write(io::Error),

and update src/uu/tsort/locales/en-US.ftl, src/uu/tsort/locales/en-FR.ftl

[xtqqczze]

ff17a13

[dragutreis]

Alright, will do, thanks for the suggestion.

[xtqqczze]

Suggested change

	/// Wrapper for bubbling up IO errors
	#[error("{0}")]
	IO(#[from] io::Error),

is it unused?

[dragutreis]

Seems like it, lemme double check 1 sec.

[dragutreis]

I may have made a mistake, I messed up my local branch state while testing the suggestion. Pushed to the PR on accident.

[xtqqczze]

at this point you probably want to rebase to fixup the history, feel free to add me as co-author :)

[dragutreis]

Good idea, thanks a lot!

[dragutreis]

Done!

[dragutreis]

My bad, will do

[xtqqczze]

nit: separate with newline (amend last commit)