convert n-dim array to pandas dataframe

I just noticed that doing

da = xr.DataArray(np.random.rand(10, 10, 10), dims=('a', 'b', 'c'))
da.to_pandas()

gives

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
~/local/bin/anaconda3/envs/test_env/lib/python3.8/site-packages/xarray/core/dataarray.py in to_pandas(self)
   2397         try:
-> 2398             constructor = constructors[self.ndim]
   2399         except KeyError:

KeyError: 3

During handling of the above exception, another exception occurred:

ValueError                                Traceback (most recent call last)
<ipython-input-22-2c9fae875981> in <module>
----> 1 da.to_pandas()

~/local/bin/anaconda3/envs/test_env/lib/python3.8/site-packages/xarray/core/dataarray.py in to_pandas(self)
   2398             constructor = constructors[self.ndim]
   2399         except KeyError:
-> 2400             raise ValueError(
   2401                 "cannot convert arrays with %s dimensions into "
   2402                 "pandas objects" % self.ndim

ValueError: cannot convert arrays with 3 dimensions into pandas objects

It is worth raising a suggestion (see below)? Or could you even do da.to_dataset(name='da').to_dataframe() under the hood. I feel the output would be what the user expects but I imagine that is not always true

da.to_dataset(name='da').to_dataframe()

Start with the DataArray.to_pandas() entry point and compare its current behavior with da.to_dataset(name='da').to_dataframe() for the three-dimensional example. First confirm whether n-dimensional conversion is intended and what output users should expect; done means an agreed behavior implemented and covered for this example.