Fms2 io update: interpolator

[uramirez8707]

Description

• Added namelist, “ngroup” to interpolator_nml which determine the number of groups to the divide the pelist to (similar to io_layout)
• ngroup = 24 will have 24 “root” ranks reading the data and broadcasting it to the other ranks
• The default is all ranks read

Fixes # (issue)

How Has This Been Tested?
ESM4.2 with 1728x2a, 1 month
Interpolator is used for ~97 files, data is read in the main loop

ngroup	Initialization	Main loop
-1	135s	580s
1	140s	595s

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published in downstream modules
• New check tests, if applicable, are included
• make distcheck passes

[bensonr]

@uramirez8707 - for this update, is each group reading the full dataset or only the relevant portions for the cores in the group? You mention you tested a 1 month run, what is the frequency of the data being ingested? If it is monthly, we may have ingested the needed timepoints on first touch or use of the data.

[uramirez8707]

@bensonr each group is reading the full dataset
I think usually the data is montly