[BUG] iloc/loc keeps circular reference to original DataFrame/Series

[mroeschke]

Describe the bug
Since loc and iloc use cached_property, the indexer classes will hang on to a reference to self even when the original object was dereferenced

Steps/Code to reproduce bug

In [1]: import weakref, cudf, pandas as pd

In [2]: df = cudf.DataFrame(range(1))

In [3]: ref = weakref.ref(df)

In [4]: df.iloc[0]
Out[4]: 
0    0
Name: 0, dtype: int64

In [5]: del df

In [6]: ref() is None
Out[6]: False

In [7]: df1 = pd.DataFrame(range(1))

In [8]: ref1 = weakref.ref(df1)

In [9]: df1.iloc[0]
Out[9]: 
0    0
Name: 0, dtype: int64

In [10]: del df1

In [11]: ref1() is None
Out[11]: True

Expected behavior
Out[6] should be True

Performance impact of just using @property (which I think is OK to eliminate this reference cycle)

In [2]: df = cudf.DataFrame(range(1))

In [3]: %timeit df.loc  # property
272 ns ± 24.3 ns per loop (mean ± std. dev. of 7 runs, 1,000,000 loops each)

In [3]: %timeit df.loc  # cached_property
46.3 ns ± 1.63 ns per loop (mean ± std. dev. of 7 runs, 10,000,000 loops each)

Environment overview (please complete the following information)

• Environment location: Bare-metal
• Method of cuDF install: conda
  • If method of install is [Docker], provide docker pull & docker run commands used

Environment details
Please run and paste the output of the cudf/print_env.sh script here, to gather any other relevant environment details

Additional context
Add any other context about the problem here.