BUG: Series.replace with value=None explicitly

[jbrockmendel]

• closes BUG: replacing a string with None causes an unexpected forward fill  #36984
• closes Series replace, unexpected fill behavior #19998
• closes df[col].dtype==object reverts back to df[col].dtype==datetime upon df.replace({np.nan: None}) #44498
• tests added / passed
• Ensure all linting tests pass, see here for how to run them
• whatsnew entry

Not sure if there is a way to keep the default values rendered in the docs unchanged.

[jreback]

i think doc-string needs updating

> /home/runner/work/pandas/pandas/pandas/core/generic.py:6801: DocTestFailure
_________________ [doctest] pandas.core.series.Series.replace __________________
5211     dtype: object
5212 
5213     When ``value=None`` and `to_replace` is a scalar, list or
5214     tuple, `replace` uses the method parameter (default 'pad') to do the
5215     replacement. So this is why the 'a' values are being replaced by 10
5216     in rows 1 and 2 and 'b' in row 4 in this case.
5217     The command ``s.replace('a', None)`` is actually equivalent to
5218     ``s.replace(to_replace='a', value=None, method='pad')``:
5219 
5220     >>> s.replace('a', None)
Differences (unified diff with -expected +actual):
    @@ -1,6 +1,6 @@
    -0    10
    -1    10
    -2    10
    -3     b
    -4     b
    +0      10
    +1    None
    +2    None
    +3       b
    +4    None
     dtype: object

[jreback]

very nice!