BUG: xarray and multi_file_day and data padding.

[rstoneback]

Describe the bug
Checks on the time index aren't consistent across pandas and xarray. Things work for pandas, they don't for xarray.

To Reproduce
Steps to reproduce the behavior:

1. gold = pysat.Instrument("ses14", "gold", "nmax")
2. gold.multi_file_day = True
3. gold.load(2020, 1)
4. See error

File ~/Code/pysat/pysat/_instrument.py:2935, in Instrument.load(self, yr, doy, end_yr, end_doy, date, end_date, fname, stop_fname, verifyPad, use_header, **kwargs)
   2933     self._prev_data.sort_index(inplace=True)
   2934 if not self._index(self._curr_data).is_monotonic_increasing:
-> 2935     self._curr_data.sort_index(inplace=True)
   2936 if not self._index(self._next_data).is_monotonic_increasing:
   2937     self._next_data.sort_index(inplace=True)

AttributeError: 'Dataset' object has no attribute 'sort_index'

The error is on _curr_data and not _prev_data since there is no data for _prev_data

Expected behavior
Gold data goes past date boundaries. Trying to clean that up with existing pysat features. Lack of testing got us here :(

Screenshots
If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS: MacOS
• Version [e.g. 22] Python 3.9
• Other details about your setup that could be relevant

Additional context
🐬

[rstoneback]

This is only a problem if the times aren't monotonic.

[rstoneback]

Short term fix, put in type check and put in correct code for xarray.

[jklenzing]

xarray.Dataset does not have the sort_index method. It needs data.sortby('time')

[jklenzing]

I've got a fix in the queue, but testing is going to be a nightmare. We currently do not test for a non-monotonic index in the padding tests. Adding the malformed_index kwarg into the tests currently has 12 broken tests, 6 passing tests per class.