Generalize weighted z-star coordinate

[cbegeman]

This PR increases the generalizability of vertical coordinates by relaxing the assumption that restingThickness spans ssh=0 to bottomDepth. As such, it is tangentially related to inactive top cells. The change affects cases where config_ALE_thickness_proportionality='restingThickness_times_weights', which is the default.

[non-BFB]

[cbegeman]

This PR is functionally similar to E3SM-Project#5254, except that formulation encountered failures with gnu compilers on chrysalis detailed here E3SM-Project#5564.

[cbegeman]

So far, this PR has small diffs on the following tests:

• SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_gnu.mpaso-jra_1958
• All IcoswISC240_WOA23 tests

[cbegeman]

@xylar As we chatted about, I'm re-opening this PR with a new formulation that seems not to be crashing on gnu compilers now. When you have a chance, take a look at the one commit f9d2573 and let me know what you think!

[cbegeman]

@xylar This case ran successfully (SMS_D_Ld1.TL319_IcoswISC30E3r5.GMPAS-JRA1p5-DIB-PISMF.chrysalis_gnu.mpaso-jra_1958) but I'm having trouble with running even master on chrysalis using a new gnu compass environment.

[xylar]

Could we call restingThicknessDiff something else (like restingSSH)? It isn't clear what it's the difference with respect to.

[xylar]

I'm having trouble with running even master on chrysalis using a new gnu compass environment.

Could you share more specifically (perhaps on Slack) what trouble you're running into? I just rebuilt my conda environment and MPAS-Ocean with gnu on Chrysalis and so far I'm having no trouble running the pr suite. Any chance you're past your quota or something like that?

Update: the pr suite finished fine:

/lcrc/group/e3sm/ac.xylar/compass_1.4/chrysalis/test_20240605/pr_gnu

Let me know what I can do to help debug the troubles you're having.