cp: fix --remove-destination deleting source when dest is the same file under a different path

[LoukasPap]

Fixes #13103

Problem

It compared the source and destination inputs as passed, so for example "a" and "./a" were compared and were -of course- found different.

Solution

We convert them to absolute paths using canonicalize(...) and then perform the equality check.

[codspeed-hq]

Merging this PR will degrade performance by 39.23%

⚠️ 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

❌ 5 regressed benchmarks
✅ 318 untouched benchmarks
⏩ 46 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	cp_recursive_deep_tree[(120, 4)]	13 ms	40.8 ms	-68.06%
❌	Simulation	cp_recursive_balanced_tree[(5, 4, 10)]	70.5 ms	102.1 ms	-30.9%
❌	Simulation	cp_preserve_metadata[(5, 4, 10)]	73.5 ms	105.5 ms	-30.32%
❌	Simulation	cp_archive_balanced_tree[(5, 4, 10)]	76.8 ms	108.3 ms	-29.11%
❌	Simulation	cp_recursive_wide_tree[(6000, 800)]	184.1 ms	242.3 ms	-24%

Tip

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

---

_{Comparing LoukasPap:issue13103 (eadcff4) with main (e73b8a6)}

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:

Skipping an intermittent issue tests/misc/tty-eof (passes in this run but fails in the 'main' branch)
Skipping an intermittent issue tests/timeout/timeout-group (passes in this run but fails in the 'main' branch)
Congrats! The gnu test tests/basenc/bounded-memory is now passing!
Congrats! The gnu test tests/rm/many-dir-entries-vs-OOM is now passing!
Congrats! The gnu test tests/tail/tail-n0f is now passing!

[LoukasPap]

Is there a way to test the performance locally or before pushing?

[LoukasPap]

Changed order of conditions to allow same-file check to happen only when --remove-destination param is set, thus avoiding redundant comparison of source and dest.