Parquet: add as_any and null_count to dyn Statistics

[ritchie46]

This adds the as_any method on dyn Statistics for parquet reading. I also added null_count similar to how dyn Statistics is defined in parquet2.

I think the distinct_count would also be valuable on this trait.

[jorgecarleitao]

Thanks! (waiting for green)

[jorgecarleitao]

fwiw, I haven't seen a parquet file with a populated distinct count xD

[codecov]

Codecov Report

Merging #743 (a6274a3) into main (0a1a878) will decrease coverage by 0.07%.
The diff coverage is 0.00%.

@@            Coverage Diff             @@
##             main     #743      +/-   ##
==========================================
- Coverage   70.79%   70.71%   -0.08%     
==========================================
  Files         313      313              
  Lines       16899    16915      +16     
==========================================
- Hits        11964    11962       -2     
- Misses       4935     4953      +18     

Impacted Files	Coverage Δ
src/io/parquet/read/statistics/binary.rs	71.42% <0.00%> (-19.49%)	⬇️
src/io/parquet/read/statistics/boolean.rs	66.66% <0.00%> (-33.34%)	⬇️
src/io/parquet/read/statistics/fixlen.rs	50.00% <0.00%> (-4.84%)	⬇️
src/io/parquet/read/statistics/mod.rs	70.83% <ø> (ø)
src/io/parquet/read/statistics/primitive.rs	59.25% <0.00%> (-10.31%)	⬇️
src/io/avro/read/schema.rs	54.83% <0.00%> (-1.08%)	⬇️
src/io/parquet/read/nested_utils.rs	77.45% <0.00%> (-0.99%)	⬇️

---

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 0a1a878...a6274a3. Read the comment docs.