Add additional ice_shelf_2d test cases

[cbegeman]

This PR adds new ice_shelf_2d test cases which are combinations of:

• 2km resolution
• single_layer vertical coordinate
• tidal_forcing version of the default test case

There are also changes made to the cfg options related to the cavity geometry which I think are more intuitive for users looking to modify the cavity geometry. In addition, the isomip_plus plotter class was imported to generate new plots.

Checklist

• User's Guide has been updated
• Developer's Guide has been updated
• N/A API documentation in the Developer's Guide (api.rst) has any new or modified class, method and/or functions listed
• Documentation has been built locally and changes look as expected
• Document (in a comment titled Testing in this PR) any testing that was used to verify the changes

[cbegeman]

Testing

Tests were run on chrysalis with intel, openmpi. I have not yet done a comparison of this branch against a compass baseline.

[xylar]

@cbegeman, I'm trying to run all the ice_shelf_2d tests but it seems like ssh_adjustment is failing for all the ones I've tried so far. I'm also trying with Chrysalis using Intel and OpenMPI so I don't know what might be different with your tests vs mine. Maybe you aren't running all the tests, just the new ones?

[xylar]

ocean/ice_shelf_2d/2km/z-star/default
  * step: initial_state
  * step: ssh_adjustment
      Failed
  test execution:      ERROR
  see: case_outputs/ocean_ice_shelf_2d_2km_z-star_default.log
  test runtime:        00:05
ocean/ice_shelf_2d/2km/z-star/tidal_forcing
  * step: initial_state
  * step: ssh_adjustment
      Failed
  test execution:      ERROR
  see: case_outputs/ocean_ice_shelf_2d_2km_z-star_tidal_forcing.log
  test runtime:        00:02
ocean/ice_shelf_2d/2km/z-star/restart_test
  * step: initial_state
  * step: ssh_adjustment
      Failed
  test execution:      ERROR
  see: case_outputs/ocean_ice_shelf_2d_2km_z-star_restart_test.log
  test runtime:        00:02
ocean/ice_shelf_2d/2km/z-level/default
  * step: initial_state
  * step: ssh_adjustment
      Failed
  test execution:      ERROR
  see: case_outputs/ocean_ice_shelf_2d_2km_z-level_default.log
  test runtime:        00:03
ocean/ice_shelf_2d/2km/z-level/tidal_forcing
  * step: initial_state
  * step: ssh_adjustment
      Failed
  test execution:      ERROR
  see: case_outputs/ocean_ice_shelf_2d_2km_z-level_tidal_forcing.log
  test runtime:        00:02
ocean/ice_shelf_2d/2km/z-level/restart_test
  * step: initial_state
  * step: ssh_adjustment
      Failed
  test execution:      ERROR
  see: case_outputs/ocean_ice_shelf_2d_2km_z-level_restart_test.log
  test runtime:        00:02
ocean/ice_shelf_2d/2km/single_layer/default
  * step: initial_state
  * step: ssh_adjustment

Compass is just hanging on the last one, though there is clearly an error message from MPAS-Ocean. I've been seeing that pretty often lately and have no idea why it hangs rather than returning with an error, but that isn't the primary problem. All ssh_adjustment steps above seem to be failing with state validation problems.

[cbegeman]

@xylar Hmm, I'll look into it.

[cbegeman]

@xylar All of the tests run now. I reduced the time step for all cases, reduced the ice shelf basal slope, and removed the inflection point in the ice shelf basal slope. I also set the threshold for deltaSSHMax to 1e-10, which only means that we reduce the number of iterations for z-level coordinates.

[xylar]

@cbegeman, these changes are really great to have! Most of my comments are just aesthetic and kind of meant to nudge things in a Polaris direction. But feel free to ignore whichever you don't think are worth fixing. We can leave some or all of this for later.

[cbegeman]

@xylar I just needed a chance to migrate your suggestions, fixup, and test. I think it's looking good now, minus the edge issues with the plot_horiz images.

[xylar]

Looks great!