PERF-#5533: Improved sort_values by reducing the number of partitions

[AndreyPavlenko]

1. In groupby_reduce() num_splits is limited by the number of partititons. We assume here, that gorupby should not increase the current data size.
2. Added a heuristic to the text file reader for calculating the number of partitions:
   num_partitions = min((num_rows * num_cols) // 64_000, NPartitions.get())
   An approximate number of rows is estimated by reading the first 10 lines.

What do these changes do?

• first commit message and PR title follow format outlined here

  NOTE: If you edit the PR title to match this format, you need to add another commit (even if it's empty) or amend your last commit for the CI job that checks the PR title to pick up the new PR title.

• passes flake8 modin/ asv_bench/benchmarks scripts/doc_checker.py
• passes black --check modin/ asv_bench/benchmarks scripts/doc_checker.py
• signed commit with git commit -s
• Resolves [PERF] Slow sort_values in value_counts #5533
• tests added and passing
• module layout described at docs/development/architecture.rst is up-to-date

[anmyachev]

@AndreyPavlenko could you provide perf results for reproducer from #5533 (comment)?

[AndreyPavlenko]

@AndreyPavlenko could you provide perf results for reproducer from #5533 (comment)?

This reproducer requires another one fix for read_csv(), that was dropped in the last commit - AndreyPavlenko@efd91bf#diff-b73ae9581d0213011834cbe1316a85876e77c2bf00b5d93e9e05be078699f04fR1090 . It was decided to implement a separate solution for read_csv().

[anmyachev]

@AndreyPavlenko could you provide perf results for reproducer from #5533 (comment)?

This reproducer requires another one fix for read_csv(), that was dropped in the last commit - AndreyPavlenko@efd91bf#diff-b73ae9581d0213011834cbe1316a85876e77c2bf00b5d93e9e05be078699f04fR1090 . It was decided to implement a separate solution for read_csv().

Should we create another issue in this case?

[AndreyPavlenko]

Should we create another issue in this case?

#6616

[anmyachev]

LGTM!