Port ice_shelf_2d test cases

[cbegeman]

This is largely a direct port of the ice_shelf_2d test cases from compass. The main changes are the following:

• Port does not include thin film and tidal forcing tests. Their utility remains unclear from my wetting-and-drying work.
• [WIP I will increase time steps to their original values] Barotropic time step has been reduced from 15s to 5s for the 5km-resolution configuration. This is not necessary, but I thought that the ratio of baroclinic:barotropic time steps should be increased from 2 to 6 to be more representative of typical configurations.
• The restart test now shares the forward step (the first run) with the default test. This is motivated mostly by computational efficiency.

Checklist

• [X] User's Guide has been updated
• [X] Developer's Guide has been updated
• [X] API documentation in the Developer's Guide (api.md) has any new or modified class, method and/or functions listed
• [X] Documentation has been built locally and changes look as expected
• [X] Testing comment in the PR documents testing used to verify the changes
• [X] New tests have been added to a test suite

[cbegeman]

@xylar This PR includes the ssh_adjustment framework I added. Please suggest changes if something different would be more convenient for isomip_plus.

[cbegeman]

@xylar Also, when you have a chance, let me know what you think of this restart test approach. I'd like to do something similar for the baroclinic channel test.

[xylar]

@cbegeman, thanks so much for this work! I'll take a look as soon as I can.

[cbegeman]

@xylar I think I've made all the changes you requested to the ssh_adjustment framework. Please let me know if there's anything else that is holding up your isomip_plus work. Otherwise, I'll move on to finishing up the rest of the tests in the group and docs and docstrings.

[xylar]

@cbegeman, I haven't had a chance to look over the latest updates today but I do think I have what I need for isomip_plus and you can finalize the docs and anything else remaining for this development when you have time.

[cbegeman]

@xylar Also, when you have a chance, let me know what you think of this restart test approach. I'd like to do something similar for the baroclinic channel test.

@xylar Whenever you have time

[xylar]

Oh, thanks! I lost track of that one.

[xylar]

The restart approach you have here looks great to me!

[cbegeman]

Testing

All new single_layer and z-star tests have been run on chrys with intel, impi.

[cbegeman]

@xylar I think this is all set to review. It took changing the number of vertical levels (from 20 to 50, though I didn't test values in between) to revive z-level (otherwise ssh_adjustment hangs).

[cbegeman]

@xylar When you have a chance, you can try the new setup_ssh_adjustment_steps approach with an ssh_forward class and an overridden compute_cell_count method.

[xylar]

@cbegeman, wow, that was fast work! I'll try this out again tomorrow.

[xylar]

@cbegeman, once you've had time to look over my latest comments, I'm happy to make a commit to this branch to make all those changes and then another one to update ice_shelf_2d with the changes. But I don't want to do that until I have the go-ahead from you.

[cbegeman]

@xylar All the changes look good to me. Please do push them; I think at this point you've done more testing than I have. Thanks for the thorough look at the code!

[cbegeman]

@xylar This is ready for your review. Thanks!

[cbegeman]

@xylar Thanks for your review and testing!