split: avoid arithmetic overflow on a huge numeric or hex suffix start value

[SAY-5]

Fixes #13749.

When split auto-computes the suffix length it adds the suffix start value (--numeric-suffixes=N / --hex-suffixes=N) to the chunk count as start as u64 + chunks. A start value near u64::MAX overflows that add and panics with attempt to add with overflow under overflow checks (exit 134).

This uses saturating_add so the auto-width calculation no longer overflows; an out-of-range start value now flows into the existing suffix-length error instead of aborting.

Added regression tests for --numeric-suffixes and --hex-suffixes start values near u64::MAX.

[codspeed-hq]

Merging this PR will improve performance by 3.76%

⚡ 2 improved benchmarks
✅ 355 untouched benchmarks
⏩ 50 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	cksum_crc32b	39.7 ms	38.1 ms	+4.19%
⚡	Simulation	cksum_sysv	68.4 ms	66.2 ms	+3.32%

Tip

Curious why this is faster? Comment @codspeedbot explain why this is faster on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing SAY-5:split-fix-suffix-start-overflow (d171ceb) with main (9875296)}

Footnotes

1. 50 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. ↩

[github-actions]

GNU testsuite comparison:

GNU test failed: tests/tail/truncate. tests/tail/truncate is passing on 'main'. Maybe you have to rebase?
Skip an intermittent issue tests/tail/overlay-headers (fails in this run but passes in the 'main' branch)
Skipping an intermittent issue tests/tail/tail-n0f (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/cp/link-heap is now passing!