Improved null count calculation (5x)

[jorgecarleitao]

The idea behind is that we do not need to merge bytes together to count zeros and ones since the position of the bit is not relevant.

git checkout d4b6aaa0985d44b9fc12a12524aa3954a8ce1537
cargo bench --no-default-features --bench count_zeros -- "2\^20"
git checkout unset_count
cargo bench --no-default-features --bench count_zeros -- "2\^20"

count_zeros 2^20        time:   [135.69 us 136.82 us 138.37 us]                             
                        change: [-82.268% -82.051% -81.805%] (p = 0.00 < 0.05)

count_zeros offset 2^20 time:   [136.00 us 136.74 us 137.59 us]                                    
                        change: [-82.323% -82.159% -81.997%] (p = 0.00 < 0.05)

Also renamed null_count to count_zeros (closes #342)

[codecov]

Codecov Report

Merging #343 (a063b99) into main (f79ae3e) will increase coverage by 0.09%.
The diff coverage is 98.03%.

@@            Coverage Diff             @@
##             main     #343      +/-   ##
==========================================
+ Coverage   80.97%   81.07%   +0.09%     
==========================================
  Files         326      326              
  Lines       21167    21357     +190     
==========================================
+ Hits        17141    17315     +174     
- Misses       4026     4042      +16     

Impacted Files	Coverage Δ
src/bitmap/immutable.rs	84.61% <66.66%> (ø)
src/bitmap/mutable.rs	89.21% <100.00%> (ø)
src/bitmap/utils/mod.rs	100.00% <100.00%> (ø)
tests/it/bitmap/utils/mod.rs	100.00% <100.00%> (ø)
tests/it/io/csv/read.rs	100.00% <0.00%> (ø)
src/array/equal/mod.rs	91.26% <0.00%> (+1.58%)	⬆️
src/io/json/read/infer_schema.rs	91.70% <0.00%> (+3.03%)	⬆️
src/io/csv/read/deserialize.rs	60.28% <0.00%> (+4.03%)	⬆️
src/io/json/read/deserialize.rs	80.55% <0.00%> (+5.75%)	⬆️

---

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 f79ae3e...a063b99. Read the comment docs.