Make whitespace descriptions visible in unicode-show

[assisted-by-ai]

Summary

• render whitespace and other non-printable characters with repr() in unicode-show descriptions so they remain visible when flagged
• add regression coverage to ensure trailing spaces and tabs are described explicitly

Testing

• python -m pytest tests/test_unicode_show.py

---

Codex Task

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Keep utility smoke tests in correct working directory

The new cd "${git_toplevel}" before running the unicode_show pytest leaves the script in the repo root, but the subsequent smoke tests still execute ./stcat/./stcatn relative to the current directory. Those binaries live under usr/bin, so ./run-tests now errors out with “No such file or directory” when it reaches the utility loops. The working directory needs to be restored to usr/bin (or the paths made absolute) after the pytest call so the smoke tests can run.

Useful? React with 👍 / 👎.

[assisted-by-ai]

Added pytest-based coverage for unicode-show, exercising ASCII control characters, Unicode format controls, trailing whitespace reporting, and clean ASCII handling.

Updated run-tests to invoke the new unicode-show tests alongside existing checks.

Treated carriage returns as suspicious characters, preserving raw line endings from files and stdin so CR bytes are no longer normalized away.

Documented the new carriage-return handling in the unicode-show manual.

Added a regression test verifying carriage returns are surfaced by unicode-show.

Ensured unicode-show treats Unicode decode failures as errors by tracking decode issues separately and exiting with status 2 when they occur.

Added a regression test that feeds invalid UTF-8 via a file to verify the tool now reports decode errors with the correct exit code.

Treated any presence of the NOCOLOR environment variable as disabling ANSI output, ensuring unicode-show stays uncolored even when the variable is set to an empty string.

Added a pseudo-terminal regression test that confirms NOCOLOR suppresses color escape codes while still reporting suspicious Unicode characters.

Treated the standard NO_COLOR environment variable the same as NOCOLOR, preventing unicode-show from emitting ANSI codes whenever either variable is set while stdout is a TTY.

Documented the new NO_COLOR handling alongside existing color-disabling conditions in the unicode-show man page.

Added a pseudo-terminal regression test confirming NO_COLOR suppresses colorized output while still flagging suspicious Unicode characters.

Ensured unicode-show renders whitespace or other non-printable characters with their repr() representation in character descriptions so flagged whitespace stays visible.

Added a regression test confirming trailing spaces and tabs are explicitly described in the output.