Speedup/simplify bitwise operations (avoid extra allocation)

[Dandandan]

Avoids an extra allocation for binary bitwise ops.

[codecov]

Codecov Report

Merging #586 (e68cda8) into main (e9d618c) will increase coverage by 0.04%.
The diff coverage is 93.10%.

@@            Coverage Diff             @@
##             main     #586      +/-   ##
==========================================
+ Coverage   78.99%   79.04%   +0.04%     
==========================================
  Files         399      399              
  Lines       24752    24743       -9     
==========================================
+ Hits        19554    19558       +4     
+ Misses       5198     5185      -13     

Impacted Files	Coverage Δ
benches/bitwise.rs	0.00% <ø> (ø)
src/bitmap/bitmap_ops.rs	86.45% <93.10%> (-2.76%)	⬇️
src/array/growable/mod.rs	21.56% <0.00%> (-4.63%)	⬇️
src/compute/arithmetics/time.rs	44.89% <0.00%> (-2.05%)	⬇️
src/bitmap/utils/slice_iterator.rs	91.04% <0.00%> (-1.50%)	⬇️
src/compute/cast/dictionary_to.rs	24.39% <0.00%> (-0.61%)	⬇️
src/io/parquet/read/mod.rs	32.62% <0.00%> (-0.37%)	⬇️
src/ffi/schema.rs	69.03% <0.00%> (-0.20%)	⬇️
src/array/mod.rs	59.64% <0.00%> (ø)
src/types/mod.rs	23.61% <0.00%> (ø)
... and 13 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update e9d618c...e68cda8. Read the comment docs.

[jorgecarleitao]

please ignore the security audit: I created the advisory for flatbuffers but have not finished auditing this and arrow-format crate to double check that we are not using any of the unsound but safe APIs incorrectly.