BUG: Make nullable booleans numeric

[dsaxton]

• closes BUG: DataFrameGroupBy.sum ignores min_count for boolean data type #34051
• tests added / passed
• passes black pandas
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

Also a decent speedup since this is now going through Cython space:

df = pd.DataFrame({"a": [1] * 50_000 + [2] * 50_000, "b": pd.array([True] * 100_000)})

# master
%timeit df.groupby("a").sum()                                                                                                                                                                  
# 5.15 ms ± 18 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)

# branch
%timeit df.groupby("a").sum()                                                                                                                                                                  
# 1.94 ms ± 10.8 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)

[pep8speaks]

Hello @dsaxton! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2020-05-10 15:59:48 UTC

[jreback]

dont use pd import anywhere, rather prefer to import directly.

[dsaxton]

I'm hitting some circular import issues when I try to import these directly, is it okay to import from within the function itself?

[jreback]

don't use pd

[jreback]

you can either import (best) or return a string "Int64"

[jreback]

thanks