Better sea ice thickness and snow depth insertion into CICE files

[shlyaeva]

It is known that soca2cice that is currently used to insert ice analysis (ice concentration, [ice thickness, snow depth]) is not working well for updating ice thickness and snow depth. Improving on this is a prerequisite for better initialization of ice thickness and snow depth.

[shlyaeva]

To improve on this, I have rewritten soca's soca2cice, following @DmitryDukhovskoy python's scripts for climatology insertion. See details in JCSDA-internal/soca#1246. Currently under testing.

@guillaumevernieres to use this, you can use:

• https://github.com/NOAA-EMC/GDASApp/tree/feature/newsoca2cice and
• https://github.com/JCSDA-internal/soca/tree/feature/new_soca2cice

Note that if you checked those out before, you'd need to update soca, and probably re-checkout gdasapp branch (I've removed the old branch, and pushed a newly created one with the same name).

I have just successfully run one cycle with 30 ensemble forecasts and one deterministic forecast, after @DmitryDukhovskoy's help with analyzing why the first attempt resulted in Icepack failing.