Add support for DataFrame(Categorical)

[michaelaye]

This would be in symmetry with the existing support of DataFrame(np.array) and DataFrame(Series).

[jreback]

In [1]: DataFrame(Series(list('aabbc')).astype('category'))
Out[1]: 
   0
0  a
1  a
2  b
3  b
4  c

In [2]: DataFrame(Series(list('aabbc')).astype('category')).dtypes
Out[2]: 
0    category
dtype: object

DataFrame(Categorical(list('aabbc')) should be equivalent of [1]