Make RWR calculations on the native grid the default

[ppharris]

In common_analysis.make_analysis() the code defaults to using wfdei.regrid_wfdei():

    ###########################################################################
    # Get a function for deriving RWR output grid info.
    ###########################################################################
    if file_sftlf is None:
        regrid_model = wfdei.regrid_wfdei
    else:
        def regrid_model(res):
            grid = read_grid(file_sftlf)
            land = [[i] for i in range(grid.nl)]
            return grid, land

which is ultimately called as the hard-wired regrid_model(1). That means the user has to do more work to get the code to calculate RWR on the native input grid of the data (e.g., 0.5 deg for observations) than to calculate it on a 1 deg coarsened version of one particular native grid (WFDEI). It seems like this behaviour is the wrong way around for general use.