[FEA] Support aggregates on Decimal columns in cudf Python

[randerzander]

In a recent cudf nightly conda package ('0.19.0a+250.g8632ca0da3'):

Example:

import cudf
from cudf.core.dtypes import Decimal64Dtype

df = cudf.DataFrame({'val': [0.01, 0.02, 0.03], 'id': [0, 1, 2]})

df['dec_val'] = df['val'].astype(Decimal64Dtype(7,2))
df['dec_val'].sum()

Result:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-37-0c210b00ba4f> in <module>
----> 1 df['dec_val'].sum()

/conda/lib/python3.8/site-packages/cudf/core/series.py in sum(self, axis, skipna, dtype, level, numeric_only, min_count, **kwargs)
   4454             )
   4455 
-> 4456         return self._column.sum(
   4457             skipna=skipna, dtype=dtype, min_count=min_count
   4458         )

/conda/lib/python3.8/site-packages/cudf/core/column/column.py in sum(self, skipna, dtype, min_count)
   1275         self, skipna: bool = None, dtype: Dtype = None, min_count: int = 0
   1276     ):
-> 1277         raise TypeError(f"cannot perform sum with type {self.dtype}")
   1278 
   1279     def product(

TypeError: cannot perform sum with type decimal