Optimal control

[axsk]

Once #29 is working it would be nice to have the pseudo optimal control $u^*$ based on $\chi$

[axsk]

@joramkuntze
This is how I parametrized/implemented this for the paper, but it is rather messy.
I first parametrize the shift-scale
https://github.com/axsk/OptImpSampling.jl/blob/97edb8b805f78fbe87d29fd43db384127562c795/src/control.jl#L126
and then compute the optimal control from the parameters
https://github.com/axsk/OptImpSampling.jl/blob/97edb8b805f78fbe87d29fd43db384127562c795/src/ociso.jl#L71

I think this would benefit greatly from a clear pen&paper representation/parametrization, as I have a hard time understanding what I did there. The main culprit should be a good parametric representation of the shift-scale.