BUG: loc.setitem modifying values with empty indexer

[phofl]

• closes BUG: loc.setitem casting dtype to float with all false Indexer and rhs is an empty list #45981 (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 type annotations to new arguments/methods/functions.
• Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

[mroeschke]

Thanks @phofl

[sam-infinity]

I have some code that broke between version 1.4.4 and 1.5.0. I was wondering if this merge will fix it. The code is essentially doing:

import pandas as pd

df = pd.DataFrame({
    "A": [],
    "B": []
}).astype('str')

df.loc[df['A'].str.contains('STUFF'), ['A']] = df['A'] + '-' + df['B']

In 1.4.4 it is a no-op but in 1.5.0 it fails with ValueError: shape mismatch: value array of shape (0,) could not be broadcast to indexing result of shape (0,1).

[phofl]

You can check your code on our nightly builds