Create an "advanced features" list

[jakobj]

It seems to me we have reached the point where some of the awesome advanced features of hal-cgp may be too hidden from the user. We should enrich the README with an overview such that users at least knows that these features exist. Maybe each of these items could then link to a documentation page or, maybe even better, an example. Let's collect all advanced features in this issue first.

Advanced features

• parallel fitness evaluation
• translation of grah into numpy/torch/sympy compatible functions
• individuals with multiple genomes/encoded functions
• hurdles
• custom operators
• local search of parameters (gradient based and evolutionary strategies)
• caching (with functional equivalence checking)
• genome reordering

Please feel free to extend this list @mschmidt87 @HenrikMettler.