DOC: GH5456 Adding workaround info on NA / NaT handling for groupby

[aamnv]

• closes DOC: update groupby NA group handing / workaround #5456
• [NA] Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• [NA] Added type annotations to new arguments/methods/functions.
• [NA] Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

Took a stab at this by parsing through the initial issue and related topics - let me know if the intent was different, happy to edit.

[mroeschke]

Thanks for the PR!

1. Could you demonstrate this by including a .. ipython:: python example (like above)
2. In the example, could you exclude inplace=True? We are trying to discourage that keyword generally

[aamnv]

Can do on both!

Do you think it makes sense to pull it out of a note and just include it in line with the rest of the text if we're including an ipython?

[mroeschke]

Yeah sure that sounds reasonable

[phofl]

I might be missing something here, but these days you can simply set dropna=False, which would include the NaNs

df = pd.DataFrame({"a": [1, np.nan, 2], "b": 3})

print(df.groupby("a", dropna=False).sum())

returns

     b
a     
1.0  3
2.0  3
NaN  3

[mroeschke]

Good point @phofl. Looks we also have an example in the groupby docstring demonstrating this behavior so I don't think this workaround explanation is needed.

Thanks for the PR @aamnv but going to close since I think this old issue is not actionable anymore, but feel free to address any of the other good first issue issues!