improve nc4 support in data_model

currently many data_model and DatasetDict tests are not carried out using netcdf4, but only xarray. this is because certain variable types (particularly time) needs extra code. I explored the code for how xarray solves this (using netcdf4 as a backend), but it is not obvious.

along these lines, the test_prms_param_separate.py does not separate parameters using netcdf4. Would be good to have that included.

Start by reading the data_model and DatasetDict tests, then inspect test_prms_param_separate.py and how the existing tests select their backend. Determine the extra handling needed for netcdf4, including time variables. Done means the relevant data_model and DatasetDict tests, plus parameter separation, also run successfully with netcdf4.

Written by the indexing model from the issue text.