cat: fix write error handling to propagate errors instead of panicking

[rynewang]

The write helper functions (write_to_end, write_tab_to_end, write_nonprint_to_end) were using .unwrap() on write operations, which would cause a panic if writing failed. This changes them to return io::Result<usize> and use ? to properly propagate errors.

The Problem

From issue #10016:

During writing, the execution result of write is captured in some places while unwrap() is directly called in others, which may trigger an abort when writing fails.

For example, in write_to_end:

// Before: panics on write error
writer.write_all(&in_buf[..p]).unwrap();

The Fix

// After: propagates error
writer.write_all(&in_buf[..p])?;

Changes

• write_to_end: returns io::Result<usize>, uses ? instead of .unwrap()
• write_tab_to_end: returns io::Result<usize>, uses ? instead of .unwrap()
• write_nonprint_to_end: returns io::Result<usize>, uses ? instead of .unwrap()
• write_end: returns io::Result<usize> to propagate errors from helpers
• Updated call site in write_lines to handle the Result with ?
• Updated unit tests to call .unwrap() on the Result

Fixes #10016

[codspeed-hq]

CodSpeed Performance Report

Merging #10038 will not alter performance

_{Comparing rynewang:fix-cat-error-handling (09f9d02) with main (81ee8d1)}

Summary

✅ 139 untouched
⏩ 37 skipped¹

Footnotes

1. 37 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. ↩