BUG: unexpected pivot after calling DataFrame.groupby('col').apply(func)

[CaioLC]

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

def generate_df() -> pd.DataFrame:
    return pd.DataFrame(
        {
            1: [1,3,5,7,9],
            2:['a', 'b', 'c', 'd', 'e'],
            'date': datetime(2023,2,1)
        },
        range(5)
    )
def generate_df2() -> pd.DataFrame:
    return pd.DataFrame(
        {
            1: [1,3,5,7,9, 10],
            2:['a', 'b', 'c', 'd', 'e', 'f'],
            'date': [datetime(2023,2,1),datetime(2023,2,1),datetime(2023,2,1),datetime(2023,2,2),datetime(2023,2,2),datetime(2023,2,2)]
        },
        range(6)
    )

def my_func(df: pd.DataFrame) -> pd.Series:
    data = [1/len(df)] * len(df)
    return pd.Series(data, df.index)

df1 = generate_df()
df2 = generate_df2()

res_unexpected = df1.groupby('date').apply(my_func) # unexpected pivoted table (index is now the column)
res2 = df2.groupby('date').apply(my_func) # this has the expected behavior. "my_func" returns a series, so the result should indeed be a series with a multiindex.

Issue Description

I have a function that accepts a dataframe and returns a pd.Series. When utilizing that inside a groupby.apply() I get different table formats depending on the GroupBy object. In the example above, if len(df['date'].unique()) == 1, the result is a DataFrame where "date" is the only index and the previous indices were pivoted into columns. If len(df['date'].unique()) > 1 I get the expected result of my custom function which is a pd.Series and the indices are 'date' and the previous index.

Expected Behavior

Output format should not vary depending on the number of unique values in a DataFrame groupby. I know that a groupby of 1 group doesn't make much sense, but it shouldn't be necessary to check that preemptively (in my specific we'll only know the unique values at runtime, so I had to add an if-else statement to handle both cases).

Installed Versions

INSTALLED VERSIONS

commit : ba1cccd
python : 3.10.11.final.0
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.19045
machine : AMD64
processor : AMD64 Family 23 Model 1 Stepping 1, AuthenticAMD
byteorder : little
LC_ALL : None
LANG : None
LOCALE : English_United States.1252

pandas : 2.1.0
numpy : 1.24.4
pytz : 2023.3
dateutil : 2.8.2
setuptools : 67.8.0
pip : 23.1.2
Cython : None
pytest : 7.4.2
hypothesis : None
sphinx : 6.2.1
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 3.1.2
IPython : 8.14.0
pandas_datareader : None
bs4 : None
bottleneck : None
dataframe-api-compat: None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : 3.7.1
numba : 0.57.1
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : None
pyreadstat : None
pyxlsb : None
s3fs : None
scipy : 1.11.2
sqlalchemy : 2.0.20
tables : None
tabulate : None
xarray : None
xlrd : None
zstandard : None
tzdata : 2023.3
qtpy : None
pyqt5 : None

[rhshadrach]

Thanks for the report! Closing as a duplicate of #54992.