[BUG] SeriesGroupBy.aggregate is inconsistent with pandas

[rjzamora]

Describe the bug
There seems to be a minor/subtle discrepancy between cuDF and Pandas for SeriesGroupBy.aggregate. If the argument to aggregate is a single string (or callable), the result should be a Series object. However, cuDF produces a DataFrame.

Steps/Code to reproduce bug

import cudf

df = cudf.DataFrame({"a": [1, 2] * 2, "b": [0] * 4})
pdf = df.to_pandas()
agg = df.groupby("a")["a"].aggregate("count")
pd_agg = pdf.groupby(["a"])["a"].aggregate("count")

assert agg.ndim == pd_agg.ndim  # Fails

cuDF result (a DataFrame):

   a
a   
1  2
2  2

Pandas result (a Series):

a
1    2
2    2
Name: a, dtype: int64

Additional context
This bug causes some test failures for dask_cudf + dask-expr in #14805 when the groupby tests are not skipped.