Improvements to wave generation - ocean waves

[mbkuhn]

Summary

I found that some cases are not stable, with problems arising from the wave generation region. This PR modifies the forcing in 2 ways to make it more reasonable and more robust.

Pull request type

• Bugfix
• Feature
• Code style update (formatting, renaming)
• Refactoring (no functional changes, no api changes)
• Build related changes
• Documentation content changes
• Other (please describe):

Checklist

The following is included:

• new unit-test(s)
• new regression test(s)
• documentation for new capability

This PR was tested by running:

• the unit tests
  • on GPU
  • on CPU
• the regression tests
  • on GPU
  • on CPU

Additional background

Creates diffs for ow_* regression tests, which are expected

[mbkuhn]

This PR also has some refactoring (introducing loops over the three components of velocity) and another bug fix (making the ramp function work as intended).

[rybchuk]

Looking forward to this! Thanks for putting this together.

[mbkuhn]

@rybchuk have you had stability issues within the forcing regions?

[rybchuk]

I've had simulations where the CFL kept blowing up after a while. I tried to get those simulations to stabilize, but never was able to. My workaround was dialing back the wave amplitude from 6.4 m (as well as 3.2 m) to 1.6 m. I figured that my code was crashing because of wave physics (are they allowed to get that big?) instead of something tied to numerics

[rybchuk]

I couldn't tell you if these simulations were crashing because of behavior in the wave input region, wave dissipation region, or elsewhere though

[mbkuhn]

Gotcha. It's very possible it's related to this.

[mbkuhn]

@marchdf Aside from the small tweaks and refactoring of things, this PR has two modifications to how the wave forcing is applied:

1. When liquid is added to a cell, assume it has the target velocity, not the current velocity of the cell
2. Avoid forcing the velocity to a target velocity when the target velocity is invalid (because the current cell is not within the target wave).

Also, I've done several things to check this manually.

[marchdf]

Sounds good! I can't run any tests right now with the downtime. We can merge and deal with the consequences or wait it out. Up to you.