fix(ptx): reject --gap-size/--width values exceeding isize::MAX

[koopatroopa787]

Fixes #13184.

Bug

ptx --gap-size N with N > isize::MAX panics (debug) or produces silent garbage / capacity overflow (release) because the output-chunk sizing math casts the value with as isize, which wraps negative on overflow:

let max_before_size =
    cmp::max(half_line_size as isize - config.gap_size as isize, 0) as usize;

GNU ptx rejects any value above isize::MAX up front:

$ ptx --gap-size 9223372036854775808 in
ptx: invalid gap width: '9223372036854775808'   # exit 1
$ ptx --width 9223372036854775808 in
ptx: invalid line width: '9223372036854775808'  # exit 1

Fix

Add two new PtxError variants (InvalidGapWidth, InvalidLineWidth) and validate the parsed usize against isize::MAX as usize immediately after parsing, before the value is stored in Config. This matches GNU's exact error messages and exit code.

Testing

Added test_gap_size_above_isize_max_is_rejected and test_line_width_above_isize_max_is_rejected. All 45 ptx tests pass; cargo clippy -D warnings and cargo fmt --check clean.

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/cut/bounded-memory (fails in this run but passes in the 'main' branch)
Skip an intermittent issue tests/date/resolution (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/date/date-locale-hour (passes in this run but fails in the 'main' branch)

[codspeed-hq]

Merging this PR will degrade performance by 4.67%

❌ 1 regressed benchmark
✅ 326 untouched benchmarks
⏩ 46 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	sort_ascii_c_locale	16 ms	16.7 ms	-4.67%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing koopatroopa787:fix-ptx-gap-size-width-overflow (a512fc8) with main (b76d615)}

Footnotes

1. 46 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. ↩