BUG: to_csv not respecting float_format for Float64

[phofl]

• closes pandas.DataFrame.to_csv float_format parameter not working as expected with pandas.Float64Dtype dtype  #45991 (Replace xxxx with the Github issue number)
• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

[jreback]

thank you @phofl

[mroeschke]

@phofl I think this is failing on main on Windows builds https://dev.azure.com/pandas-dev/pandas/_build/results?buildId=74146&view=logs&jobId=2b18e8df-2ee7-5e01-73e9-0533d995635b&j=2b18e8df-2ee7-5e01-73e9-0533d995635b&t=13dd5051-9e2f-5a32-09fe-1d338e0833eb

_________________ TestToCSV.test_to_csv_float_ea_float_format _________________
[gw1] win32 -- Python 3.10.2 C:\Miniconda\envs\pandas-dev\python.exe

self = <pandas.tests.io.formats.test_to_csv.TestToCSV object at 0x0000025FCC572B00>

        def test_to_csv_float_ea_float_format(self):
            # GH#45991
            df = DataFrame({"a": [1.1, 2.02, pd.NA, 6.000006], "b": "c"})
            df["a"] = df["a"].astype("Float64")
            result = df.to_csv(index=False, float_format="%.5f")
            expected = """a,b
    1.10000,c
    2.02000,c
    ,c
    6.00001,c
    """
>           assert result == expected
E           AssertionError: assert 'a,b\r\n1.100...6.00001,c\r\n' == 'a,b\n1.10000...\n6.00001,c\n'
E             - a,b
E             + a,b

E             ?    +
E             - 1.10000,c
E             + 1.10000,c

E             ?          +
E             - 2.02000,c...
E             
E             ?      +
E             - 2.02,c...
E             
E             ...Full output truncated (9 lines hidden), use '-vv' to show

[phofl]

Thx @mroeschke, opened #46024