Correcting HEC-RAS Boundary Errors

[sjordan29]

Discovered while addressing #70- there appears to be a bug in HEC-RAS where for certain boundaries (in our instance, the cold creek), boundary cells / faces are incorrectly attributed to those boundaries in one place in the HEC-RAS output, but correctly in others. Jason tried many things to fix it on the HEC-RAS side, so we're implementing a basic fix in

This was interfering with the postprocessing utilities in clearwater riverine and producing incorrect mass balance results. Only occurring in the fine mesh.