[BUG] Comparison between Series and DataFrame fails when LHS is a Series

[brandon-b-miller]

Describe the bug
Comparisons (eq, neq, ge, etc) fail between a Series and a DataFrame when the series is the left operand.

Steps/Code to reproduce bug

lhs = cudf.Series([1,2,3])
rhs = cudf.DataFrame({'a': [1,2,3]})
lhs == rhs

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/nfs/brmiller/repos/cudf/python/cudf/cudf/core/series.py", line 1611, in __eq__
    return self._binaryop(other, "eq")
  File "/home/nfs/brmiller/anaconda3/envs/cudf_dev/lib/python3.7/contextlib.py", line 74, in inner
    return func(*args, **kwds)
  File "/home/nfs/brmiller/repos/cudf/python/cudf/cudf/core/series.py", line 1187, in _binaryop
    return other._apply_op(self, fn)
  File "/home/nfs/brmiller/repos/cudf/python/cudf/cudf/core/dataframe.py", line 1359, in _apply_op
    result[col] = getattr(self[col], fn)(other[k])
TypeError: getattr(): attribute name must be string

It's supposed to be tested here but it currently isn't because apparently I broke it 15 months ago.

Expected behavior

       a      0      1      2
0  False  False  False  False
1  False  False  False  False
2  False  False  False  False

Environment overview (please complete the following information)

• Environment location: Bare-metal
• Method of cuDF install: Source