cumulate+argmax uses padded index instead of absolute index

What happened?

The combination of cumulative + argmax does not lead to the (at least from me) expected result.
Example:

import numpy as np
import xarray as xr
da = xr.DataArray([1,2,1.5,3.5,4], coords={'time': ('time', np.arange(5))})
da.cumulative('time').argmax()
# [4., 4., 3., 4., 4.]
# expected [0, 1, 1,3,4]

If I want to fetch the time for the largest value until that value I would use something like above cumulative('time').argmax(). However it doesnt not return the index but seems to use some kind of padded rolling array and fetches the argmax from this. Maybe this is the expected behaviour but I was definitely suprised.

MVCE confirmation

• Minimal example — the example is as focused as reasonably possible to demonstrate the underlying issue in xarray.
• Complete example — the example is self-contained, including all data and the text of any traceback.
• Verifiable example — the example copy & pastes into an IPython prompt or Binder notebook, returning the result.
• New issue — a search of GitHub Issues suggests this is not a duplicate.
• Recent environment — the issue occurs with the latest version of xarray and its dependencies.

Environment

Start by running the minimal Python example and tracing the cumulative and argmax entry points that produce the padded indices. Verify the behavior against the expected absolute indices and add regression coverage showing the cumulative argmax result for this example.