[feature request] __iter__() for rolling-window on datasets

Currently, rolling() on a dataset does not return an iterator:

MCVE Code Sample

arr = xr.DataArray(np.arange(0, 7.5, 0.5).reshape(3, 5),
    dims=('x', 'y'))

r = arr.to_dataset(name="test").rolling(y=3)
for label, arr_window in r:
    print(label)

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-12-b1703cb71c1e> in <module>
      3 
      4 r = arr.to_dataset(name="test").rolling(y=3)
----> 5 for label, arr_window in r:
      6     print(label)

TypeError: 'DatasetRolling' object is not iterable

Output of xr.show_versions()

Start with the Dataset.rolling() entry point and the DatasetRolling object shown in the MCVE, then reproduce the example in a development environment. Done means iterating over the rolling result works and yields the label and window demonstrated by the requested loop; add focused coverage for that behavior.