use existing permittivity for the design region initialization spec

[tylerflex]

as discussed, would be cool if we could evaluate the simulation.epsilon() at the design region to get some permittivity distribution to use as the starting value for invdes optimization.

The only way I can think of doing this is by first solving an optimization problem to come up with the params0 that give rise to a eps close to the desired pattern. I think we do something similar in the level set notebook to get the starting parameters for the splitter.