[Dev] Add InfiniteOpt version of the problems

[ocots]

Add an InfiniteOpt version of the problems. See here for some examples.

@jbcaillau ok with that?

[jbcaillau]

yes, though not top priority i would say. InfiniteOpt is a nice layer above JuMP, thought as an extension of the JuMP (similar syntax, etc.) since we already (should) have translations into JuMP models of the problems to be tested, i do not think it is a priority to also add InfiniteOpt models (we will gain neither performance wrt. to hand coded JuMP models, nor benchmarking information, so).

[jbcaillau]

@ocots not a priority right now, i'd say. largely enough to offer a comparison with JuMP hand-coded discretised problems. close if you want!

[ocots]

This idea was to have the possibility to compare different discretisation schemes. But I agree that it is not the top priority.

[jbcaillau]

we can actually compare different schemes in OptimalControl.jl (@PierreMartinon). closing