Switch the mesh used for Omega CTests

[xylar]

This merge updates the mesh used in Omega's CTests to an oQU240 initial condition taken from inputdata on the E3SM server. This is necessary for the HORZMESH_TEST to pass, since that test assumes realistic total ocean area and coriolis parameter.

This merge also updates the Omega submodule to the latest develop (hash: ca19fa9)

Checklist

• Testing comment in the PR documents testing used to verify the changes

[xylar]

Testing

I was able to successfully run Omega's CTests with these updates, whereas the HORZMESH_TEST was failing with the previous (Icos480 cosine-bell) horizontal mesh.

[xylar]

@sbrus89, could you have a look at this when you have time?

[xylar]

@sbrus89, still hoping you can have a look at this at some point.

[sbrus89]

Sorry @xlyar, I lost track of this. I'll finish reviewing tomorrow.

[sbrus89]

Approved based on successful testing on pm-cpu. Thanks for this fix, @xylar.

[xylar]

Thanks, @sbrus89!