[BUG] factorize returning incorrect results for CategoricalDtype

[galipremsagar]

Describe the bug
When factorize is called with a CategoricalDtype, it always returns nulls.

Steps/Code to reproduce bug

In [1]: import cudf

In [2]: s = cudf.Series(['a', 'b', 'c', 'd'])

In [4]: s.factorize()
Out[4]: 
(array([0, 1, 2, 3], dtype=int8),
 StringIndex(['a' 'b' 'c' 'd'], dtype='object'))

In [5]: s = s.astype('category')

In [6]: s.factorize()
Out[6]: 
(array([-1, -1, -1, -1], dtype=int8),
 CategoricalIndex([<NA>, <NA>, <NA>, <NA>], categories=['a', 'b', 'c', 'd'], ordered=False, dtype='category'))

In [7]: s.to_pandas().factorize()
Out[7]: 
(array([0, 1, 2, 3]),
 CategoricalIndex(['a', 'b', 'c', 'd'], categories=['a', 'b', 'c', 'd'], ordered=False, dtype='category'))

Expected behavior

In [7]: s.factorize()
Out[7]: 
(array([0, 1, 2, 3]),
 CategoricalIndex(['a', 'b', 'c', 'd'], categories=['a', 'b', 'c', 'd'], ordered=False, dtype='category'))

Environment overview (please complete the following information)

• Environment location: [Bare-metal]
• Method of cuDF install: [from source]