Over None gives None instead of str

[MariusMerkleQC]

Checks

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

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

Reproducible example

import polars as pl

df = pl.DataFrame(data = {"color": ["yellow", "yellow"], "color2": [None, "light"], "val": ["2", "3"]})

df = df.with_columns(pl.min("val").over(["color", "color2"]).alias("min_val_per_color"))
print(df)

Log output

No response

Issue description

When the column "val" is of type pl.Utf8, it will set the aggregated value to None if any value in the group is `None. You get the additional column "min_val_per_color: [None, 3]".

Expected behavior

I would expect the dataframe to get the additional column "min_val_per_color: [2, 3]". This is also what happens when the datatype of "val" is numeric.

Installed versions

--------Version info---------
Polars:               0.20.1
Index type:           UInt32
Platform:             macOS-14.1.1-arm64-arm-64bit
Python:               3.12.0 | packaged by conda-forge | (main, Oct  3 2023, 08:36:57) [Clang 15.0.7 ]

----Optional dependencies----
adbc_driver_manager:  <not installed>
cloudpickle:          <not installed>
connectorx:           <not installed>
deltalake:            <not installed>
fsspec:               <not installed>
gevent:               <not installed>
matplotlib:           <not installed>
numpy:                1.26.2
openpyxl:             <not installed>
pandas:               <not installed>
pyarrow:              <not installed>
pydantic:             <not installed>
pyiceberg:            <not installed>
pyxlsb:               <not installed>
sqlalchemy:           <not installed>
xlsx2csv:             <not installed>
xlsxwriter:           <not installed>
None

[cmdlineluser]

This last worked in 0.19.19

Seems to have changed after 34330ec