Make procedure for throughput testing of FATES runs

[samsrabin]

It would be good to have some FATES compsets in the CESM3 timing table, but the usual PFS test setup (20 days) isn't representative for runs where the number of patches can change—i.e., setups with use_fates_sp and use_fates_nocomp both false.

[samsrabin]

May block completion of #2722, if we want to include such compsets in the CESM3 timing table.

[ekluzek]

The real thing here is #1277 we need some spunup initial conditions for FATES. So maybe the central thing here is a spinup procedure for FATES that gives us initial condition files to do a PFS test with?

[samsrabin]

Good point. I've added that to the cesm3.0.0 milestone.