dd: Avoid malloc & 0 fill

[oech3]

dd_copy_8k_blocks +1%
dd_copy_with_seek +1%
dd_copy_4k_blocks +1%
dd_copy_default +1%
dd_copy_64k_blocks 601 µs -> 599.7 µs
dd_copy_partial 550.1 µs -> 550.3 µs (worse)

[codspeed-hq]

Merging this PR will not alter performance

✅ 9 untouched benchmarks
⏩ 319 skipped benchmarks¹

---

_{Comparing oech3:dd-resize (0a7884a) with main (e0694a5)}

Footnotes

1. 319 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. ↩

[oech3]

Still cannot catch perf gain similar with mimalloc.

[xtqqczze]

@oech3 I don’t believe this change achieves the behavior described in the title.

[oech3]

Avoid truncate?

[xtqqczze]

Why do we change from BUF_INIT_BYTE (0xDD) fill to zero fill?