[FEA] cudf.pivot_table aggfunc accept list of function

**Is your feature request related to a problem? Please describe.**
I wish cudf.pivot_table aggfunc could accept list of function like below
```
In [31]: df = pd.DataFrame(
...: {
...: "k1": [0, 0, 0, 0, 0, 1, 1, 1, 1, 1],
...: "k2": ['a', 'b', 'a', 'b', 'a', 'b', 'a', 'b', 'a', 'b'],
...: "c": ['c0', 'c0', 'c0', 'c1', 'c1', 'c0', 'c1', 'c1', 'c1', 'c1'],
...: "v": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9],
...: }
...: )

In [32]: df
Out[32]:
k1 k2 c v
0 0 a c0 0
1 0 b c0 1
2 0 a c0 2
3 0 b c1 3
4 0 a c1 4
5 1 b c0 5
6 1 a c1 6
7 1 b c1 7
8 1 a c1 8
9 1 b c1 9

In [33]: pd.pivot_table(df, index=["k1", "k2"], columns=['c'], values=['v'], aggfunc=['sum', 'mean'])
...:
Out[33]:
sum mean
v v
c c0 c1 c0 c1
k1 k2
0 a 2.0 4.0 1.0 4.0
b 1.0 3.0 1.0 3.0
1 a NaN 14.0 NaN 7.0
b 5.0 16.0 5.0 8.0
```

So far cudf returns this error
`ValueError: 2 columns passed, passed data had 3 columns`