[Data] Method to iterate over GroupedData

[dominikgrygiel]

Description

It would be great to have something like GroupedData.iter_groups to iterate over the groups after groupby operation.

This might be a duplicate of: #42228 I feel like we are trying to have similar functionality, but with a different approach.

Use case

Similarly to: #42228 I would like to perform some extra operations on the grouped data and persist it to disk/database. Currently, I'm doing something like the following, but it feels "wrong":

def _do_something(group: pd.DataFrame) -> pd.DataFrame:
    # Do some operations, persist to disk/database, etc.

    return group.head(0)  # Not sure if that is actually helping or would returning `group` be better

ds.groupby("column").map_groups(_do_something).materialize()