With the changes we've made to how we refer to numpy types, we need to change lines like this in test_plotting.py:
check(assert_type(df.hist(), npt.NDArray[np.object_]), npt.NDArray[np.object_])
to
check(assert_type(df.hist(), npt.NDArray[np.object_]), np.ndarray)
See numpy/numpy#30319 (comment) and comments below for discussion. We shouldn't be passing the typing type npt.NDArray as the klass to check()