od: improve perf for --skip-bytes non-seekable input by splice(2)

[oech3]

Closes #12598
This PR adds splice fast-path for --skip-bytes for non-seekable input as same as wc -c.

> truncate -s 1GB /tmp/huge
> time cat /tmp/huge|target/release/od-splice --skip-bytes 1GB
7346545000

________________________________________________________
Executed in   17.97 millis    fish           external
   usr time    0.95 millis    0.00 millis    0.95 millis
   sys time   31.25 millis    3.02 millis   28.23 millis

> time cat /tmp/huge|target/release/od --skip-bytes 1GB
7346545000

________________________________________________________
Executed in  160.61 millis    fish           external
   usr time  109.92 millis    0.89 millis  109.03 millis
   sys time  181.50 millis    2.76 millis  178.73 millis

> hyperfine "cat /tmp/huge|target/release/od --skip-bytes 1GB" "cat /tmp/huge|target/release/od-splice --skip-bytes 1GB" 
Benchmark 1: cat /tmp/huge|target/release/od --skip-bytes 1GB
  Time (mean ± σ):     155.4 ms ±  26.3 ms    [User: 105.1 ms, System: 183.9 ms]
  Range (min … max):   141.6 ms … 260.3 ms    20 runs
 
Benchmark 2: cat /tmp/huge|target/release/od-splice --skip-bytes 1GB
  Time (mean ± σ):      18.9 ms ±   6.1 ms    [User: 3.6 ms, System: 27.3 ms]
  Range (min … max):    15.5 ms …  68.1 ms    154 runs
 
Summary
  cat /tmp/huge|target/release/od-splice --skip-bytes 1GB ran
    8.24 ± 3.00 times faster than cat /tmp/huge|target/release/od --skip-bytes 1GB

[codspeed-hq]

Merging this PR will improve performance by 4.65%

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 2 improved benchmarks
✅ 345 untouched benchmarks
⏩ 46 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	du_wide_tree[(5000, 500)]	20.1 ms	19.2 ms	+4.71%
⚡	Simulation	du_max_depth_balanced_tree[(6, 4, 10)]	62.8 ms	60 ms	+4.59%

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 oech3:splice-discard (8666570) with main (371b24a)}

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. ↩

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/date/date-locale-hour (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/tail/inotify-dir-recreate (passes in this run but fails in the 'main' branch)
Note: The gnu test tests/misc/write-errors was skipped on 'main' but is now failing.