Series.apply() should accept `None` as Callable return type

[chetanpepper]

Describe the bug
Pandas is fine if the Callable passed to series.apply(...) returns None, but mypy throws an error. Passing a null-returning Callable to df.apply() does not throw an error.

To Reproduce

import pandas as pd

s = pd.Series([{'a': 1, 'b':2}])
s['c'] = s.apply(lambda x: None)

mypy fails with

a.py:4: error: Argument 1 to "apply" of "Series" has incompatible type "Callable[[Any], None]"; expected "Callable[..., Union[str, bytes, date, datetime, timedelta, datetime64, timedelta64, bool, int, float, Timestamp, Timedelta, complex, Sequence[Any], Mapping[Any, Any]]]"  [arg-type]
a.py:4: error: Incompatible return value type (got "None", expected "Union[str, bytes, date, datetime, timedelta, datetime64, timedelta64, bool, int, float, Timestamp, Timedelta, complex, Sequence[Any], Mapping[Any, Any]]")  [return-value]
Found 2 errors in 1 file (checked 1 source file)

Please complete the following information:

• MacOS
• 12.6
• python3.10
• mypy 1.0.1
• pandas-stubs 1.5.3.230227

Additional context
Using a data frame instead of a series results in no error:

import pandas as pd

df = pd.DataFrame([{'a': 1, 'b':2}])
df['c'] = df.apply(lambda x: None)

[Dr-Irv]

Thanks for the report. At least with pyright, I am seeing this as a problem for both Series.apply() and DataFrame.apply().

In the case of Series.apply(), just have to change

pandas-stubs/pandas-stubs/core/series.pyi

Line 764 in d2faa7f

func: Callable[..., Scalar | Sequence | Mapping],

to include None as a possible result in Callable.

For DataFrame.apply(), have to duplicate lines that have S1 as an argument to another overload that returns Series instead of Series[S1], as appropriate.

Tests needed for both cases.