File written by hist_printflds is missing some fields

[billsacks]

Brief summary of bug

I noticed that the file written by the hist_printflds routine (e.g., master_list_nofates.rst) is missing some fields.

General bug information

CTSM version you are using: ctsm5.1.dev057

Does this bug cause significantly incorrect results in the model's science? NO

Configurations affected: ??

Details of bug

I noticed that all of the PCT_* fields (defined in subgridWeightsMod) are missing from https://github.com/ESCOMP/CTSM/blob/master/doc/source/users_guide/setting-up-and-running-a-case/master_list_nofates.rst and from the file in the test, ERP_P36x2_D_Ld3.f10_f10_mg37.I1850Clm50BgcCrop.cheyenne_gnu.clm-extra_outputs, even though they are present in the history file for that test. I haven't looked to see if any other fields may be affected.

The problem is that hist_printflds is called a little too early in initialization – before init_subgrid_weights_mod and possibly before other relevant subroutines. This could probably be safely moved to the end of initialize2.

We should move that routine, then regenerate the documentation from these history field lists.