DOC: Mention that boolean indexing is impossible for new nullable integer/boolean and string data types when they contain missing values

[tdpetrou]

The new data types nullable integer, nullable boolean, and string cannot do boolean indexing when the columns contain missing values.

I searched the pd.NA and the boolean indexing sections and have not seen this mentioned. Users need to be aware of this limitation as it greatly reduces their usability.

[jorisvandenbossche]

The docs can certainly be better organised and interlinked, but it is (briefly) mentioned here: https://pandas.pydata.org/docs/user_guide/boolean.html#indexing-with-na-values

[anant4299]

I will add that to the pd.NA and boolean indexing sections and will search for any other section that might need that information.

[rhshadrach]

This is no longer the case - Boolean indexing is allow, and is documented properly in the link above. Closing.