[BUG] Index.isna should return an array, not an Index

[vyasr]

Describe the bug Currently, calling isna (or isnull) on any index type will return an Int64Index of dtype bool. It should instead return a cupy array to match pandas behavior (returning a numpy array).

Steps/Code to reproduce bug

>>> cudf.Index([1, 2, 3, 4, 5]).isna()
Int64Index([False, False, False, False, False], dtype='bool')

Expected behavior

>>> cudf.Index([1, 2, 3, 4, 5]).isna()
array([False, False, False, False, False])

[github-actions[bot]]

This issue has been labeled inactive-30d due to no recent activity in the past 30 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed. This issue will be labeled inactive-90d if there is no activity in the next 60 days.

[github-actions[bot]]

This issue has been labeled inactive-30d due to no recent activity in the past 30 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed. This issue will be labeled inactive-90d if there is no activity in the next 60 days.