FIX-#5925: Put a sorting-hack into groupby tests to hide #6875 bug

[dchigarev]

What do these changes do?

CI on master started to fail after #6862 was merged (https://github.com/modin-project/modin/actions/runs/7701338108/job/20995335180). The tests are failing because the order of rows doesn't match between modin and pandas. The test fails even if I won't cast the data to categoricals, so I'm blaming #6875 for the failures. The failure isn't stable (fails not always), that's why it wasn't caught before merging #6862.

My current plan is:

1. Merge this hack in order to unblock CI
2. Make the investigation of BUG: groupby(sort=True) may produce unsorted results with range-partitioning implementation #6875 my first priority

• 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 Reshuffling groupby doesn't handle grouping on a categorical column correctly #5925
• tests added and passing
• module layout described at docs/development/architecture.rst is up-to-date