[FEA] Run `get_dummies` on cudf Series

[taureandyernv]

Is your feature request related to a problem? Please describe.
I'd like to run get_dummies on a Series. When trying to perform .get_dummies on a Series, cudf throws an Attribute Error: 'Series' object has no attribute 'select_dtypes'.

Describe the solution you'd like

import pandas as pd
import cudf
import numpy

df = cudf.DataFrame()
df['a']= numpy.random.randint(0,5,20)
df['b']= numpy.random.randint(0,5,20)
df['c']= numpy.random.randint(0,5,20)
pdf = df.to_pandas()

print(pdf.head())
print(df.head())

pdf2 = pd.get_dummies(pdf['a'], prefix=['a_']) #this works
print(pdf2.head())
df2 = cudf.get_dummies(df['a'], prefix=['a_']) #this does NOT work currently
print(df2.head())

I'd like print(df2.head())'s output to be something like:

   ['a_']_0  ['a_']_1  ['a_']_2  ['a_']_3  ['a_']_4
0         0         1         0         0         0
1         0         0         0         1         0
2         0         0         0         1         0
3         1         0         0         0         0
4         0         0         0         1         0

instead of

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-30-d45df2a33ea1> in <module>
     14 pdf2 = pd.get_dummies(pdf['a'], prefix=['a_']) #this works
     15 print(pdf2.head())
---> 16 df2 = cudf.get_dummies(df['a'], prefix=['a_']) #this does NOT work
     17 print(df2.head())

/opt/conda-environments/rapids-stable/lib/python3.8/site-packages/cudf/core/reshape.py in get_dummies(df, prefix, prefix_sep, dummy_na, columns, cats, sparse, drop_first, dtype)
    564 
    565     if columns is None or len(columns) == 0:
--> 566         columns = df.select_dtypes(include=encode_fallback_dtypes).columns
    567 
    568     def length_check(obj, name):

AttributeError: 'Series' object has no attribute 'select_dtypes'

Describe alternatives you've considered
Only current alternative is to convert to pandas