Idealized time-series representing prescribed atmospheric state

[glwagner]

To motivate fleshing out the user-interface for the ThermodynamicSeaIce model, we'd like to develop an idealized time-dependent scenario representing ice growth/melting (with time-evolving atmospheric state / radiation conditions, and time-evolving ocean state).

[glwagner]

PR #7 created a time-series representing a prescribed ice surface state. Next we need to prescribe the atmosphere state (and have some simple way of computing air-ice fluxes, for example using Monin-Obukhov theory or some simplified version of it).

I think we also want to prescribe radiative fluxes as well?