Incorporating slope orientation (aspect) as an HRU variable

[Musselman]

Accurate slope-scale simulations in complex terrain require terrain orientation to be explicitly considered in the calculation of solar insolation. I've spoken to Marytn about this issue. It seems that aspect is currently a hardwired variable in engine/sunGeomtry.f90 and engine/derivforce.f90 (specified as 0) and should be brought to the front end and incorporated as an HRU-specific variable similar to slope or LAI.