ENH: Consistency of RadCLss Dataset

[jrobrien91]

Following initial data review, corrections to the RadClss processing have been made.

• within bnf_radclss.py, an prior assumption to interpolate each individual field within a datastream the same way led to values that were unphysical, particularly within the MET fields. This was updated to check for the individual fields that require summation (accumulating precipitations) and assume remaining fields can be averaged.
• there was also an invalid assumption made within the WXT variable names when applying the prefix that created scenarios where all values were MVC that were corrected
• notebook comparing RadCLss to input datastreams created