Include pressure gradient force into time-avg output

[tomchor]

This branch uses an Oceananigans branch where I fixed the time stepping issue that led to weird values in pressure. It was the only way I could get rid if this.

A note: while I do have a branch with a fix for the most up-to-date Oceananigans code in CliMA/Oceananigans.jl#3606, upgrading the Oeananigans version to the latest one in this repo breaks the code since Casper doesn't have Julia 1.10, which is necessary. So I just fixed the pressure issue separately in
a branch that branched off of v0.90.6, which is the version we're currently using in this code.

[tomchor]

Just some ugly pictures for future record, the PGF used to be quantified as this:

(The top panels are instantaneous snapshots, bottom left are time-averages, and bottom right is a time evolution of the STD of pressure, showcasing how we got errors occasionally.)

With the fix, the same results now are:

Which to me seems like there's no error at all. I think with this we might be able to close the energy budget, which will be the next PR.