RIFG, RUNW, GUNW - Use of Polyfitting leads to layers containing all zero values

@xhuang-jpl Please confirm that this is the correct description of the issue:

Issue

When polyfitting is enabled in the InSAR runconfig (via this parameter), then the correlationSurfacePeak layer(s) in the output RIFG, RUNW, and GUNW products are populated with all zero values.

This is not correct; all HDF5 science datasets (aka raster layers) should instead be populated with valid, accurate, meaningful science values.

Extent of the Issue

For products produced via NISAR Mission Operations, this issue impacts all granules produced using (1) solid earth runconfig parameters, and (2) urgent response runconfig parameters.

Action Items

1. InSAR team should determine a path forward algorithmically.
   • Goal: When Polyfitting is enabled, the correlationSurfacePeak layer(s) should be populated with valid, meaningful values.
   • The fixes could include updates to ISCE3, updates to the parameters, etc.
2. Once the issue is resolved in ISCE3, NISAR ADT should update the QA parameter set files so that nisarqa will raise an Exception and fail when the correlationSurfacePeak layer(s) contain all zeros.
   • Currently in mission operations, QA is being "quieted" and HDF5 science products containing this issue are allowed to be released.
   • The fix for QA (in brief): Reopen and merge this PR: https://github-fn.jpl.nasa.gov/NISAR-ADT/ParameterSetFiles/pull/95/files