Fix convert_dtype complex

[Ma-gi-cian]

• closes BUG: allow complex type in convert_dtypes #60129
• 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/v3.0.0.rst file if fixing a bug or adding a new feature.

Changes :

• Modified pandas/core/dtypes/cast.py to preserve complex dtypes (kind 'c') instead of attempting to convert them
• Since there is no Numpy-nullable complex array nor a PyArrow complex dtype, this is effectively a no-op (leaves complex data as-is)
• Removed 'c' (complex) from dtype kind checks where it was incorrectly grouped with floats
• Added tests for both DataFrame.convert_dtypes() and Series.convert_dtypes() with complex data

Example script and output :

import pandas as pd
print(pd.DataFrame(['1.0+5j', '1.5-3j']).astype(complex).convert_dtypes())

Output :

(pandas-dev) jha@MAGICIAN:~/workspace/opensource/reports/pandas-60129$ python3 main.py
          0
0  1.0+5.0j
1  1.5-3.0j