BUG: Series.combine_first() changes int dtypes, but not Int (extension) dtypes

[Dr-Irv]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

>>> import pandas as pd
>>> s1 = pd.Series([1, 2, 3], index=["A", "B", "C"])
>>> s2 = pd.Series([4, 6], index=["A", "C"])
>>> s1
A    1
B    2
C    3
dtype: int64
>>> s2
A    4
C    6
dtype: int64
>>> s2.combine_first(s1)
A    4.0
B    2.0
C    6.0
dtype: float64
>>> s2.convert_dtypes()
A    4
C    6
dtype: Int64
>>> s2.convert_dtypes().combine_first(s1.convert_dtypes())
A    4
B    2
C    6
dtype: Int64

Issue Description

In the example, the dtypes of the Series are int64 but when you do combine_first, the result has a dtype of float64.

If you use convert_dtypes() to change the types to Int64, then the dtype is preserved.

Expected Behavior

The int64 dtype should be preserved.

Installed Versions

show_details() didn't work!

pandas 2.0.2

[phofl]

Pretty sure there is a reindex happening under the hood which coerces the dtype. Investigations welcome

[lukemanley]

This is fixed on main (#51777). I think this can be closed.

[Dr-Irv]

This is fixed on main (#51777). I think this can be closed.

Thanks @lukemanley . I agree.