speedup nested_columns logic

[dougbrn]

Inspired by #379 , should be ~20x faster according to my testing

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.77%. Comparing base (3d3b6e0) to head (8170e86).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #380      +/-   ##
==========================================
- Coverage   97.77%   97.77%   -0.01%     
==========================================
  Files          19       19              
  Lines        2026     2024       -2     
==========================================
- Hits         1981     1979       -2     
  Misses         45       45              

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[smcguire-cmu]

Looks great, thanks!

[github-actions]

Before [3d3b6e0]	After [de6fce3]	Ratio	Benchmark (Parameter)
917M	958M	1.04	benchmarks.ReadFewColumnsS3.peakmem_run
30.7±0.6ms	31.4±0.9ms	1.02	benchmarks.AssignSingleDfToNestedSeries.time_run
64.2±0.07ms	64.5±0.5ms	1.01	benchmarks.CountNestedBy.time_run
1.56±0.01s	1.57±0.01s	1.01	benchmarks.ReadFewColumnsHTTPS.time_run
2.10±0.02s	2.13±0.05s	1.01	benchmarks.ReadFewColumnsS3.time_run
267M	270M	1.01	benchmarks.ReassignHalfOfNestedSeries.peakmem_run
43.0±0.3ms	43.3±0.6ms	1.01	benchmarks.ReassignHalfOfNestedSeries.time_run
248M	247M	1	benchmarks.AssignSingleDfToNestedSeries.peakmem_run
136M	136M	1	benchmarks.CountNestedBy.peakmem_run
102M	102M	1	benchmarks.NestedFrameAddNested.peakmem_run

Click here to view all benchmarks.