[BUG] cudf.to_numeric does not preserve the index

[rjzamora]

Describe the bug
When pd.to_numeric(<series>) is called on a series with a non-default index, that index is preserved. When cudf.to_numeric is used on the same data, the original index is replaced with a default RangeIndex

Steps/Code to reproduce bug

import cudf, pandas as pd

ser = pd.Series(["1"] * 8, index=range(2, 10))
pd.to_numeric(ser).index  # RangeIndex(start=2, stop=10, step=1)

cudf.to_numeric(ser).index  # RangeIndex(start=0, stop=8, step=1)

gser = cudf.from_pandas(ser)
cudf.to_numeric(ser).index  # RangeIndex(start=0, stop=8, step=1)

Expected behavior
I would not expect cudf.to_numeric to modify the index of the original Series at all. Even if the original index is non-numeric, it should be preserved.