cp: reduce memory usage for cp -al by skipping unnecessary tracking

[ChrisDryden]

This test makes sure that in a large folder, the memory usage doesn't go up a bunch in this mode. The change itself to fix this is just to add a if check that if its this mode to not store all of the additional metadata since there's no need to use it later.

[codspeed-hq]

CodSpeed Performance Report

Merging #9805 will not alter performance

_{Comparing ChrisDryden:fix-cp-link-heap-memory (77a16f1) with main (2d718f2)}

Summary

✅ 127 untouched
⏩ 6 skipped¹

Footnotes

1. 6 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:

Congrats! The gnu test tests/cp/link-heap is no longer failing!

[ChrisDryden]

As a note for this one, we need to come up with a framework for testing memory usage similar to codspeed so that we can verify that we don't introduce regressions that use more memory than required. I don't think codspeed currently supports it but it looks like its on its roadmap to be supported soon