fix: properly evaluate nsga population

The full population has to be evaluated at once,
offsprings cannot be evaluated individually.

Lots of changes I lost track of

fix: implement possibility of best candidate not being found

refactor: add possibility to use different logic for obtaining best candidate

feat: add possibility to map EvolutionResult evaluations

refactor(lib): Allow fitting whole population at once

The fitness function gets fit_population function. The default
implementation is going over fit() one by one, but it can be
reimplemented by specific types.

feat(lib): allow modifying fitness function in evolution_algorithm

chore: move population structs to separate module

feat: add a lot of algorithm showcases

feat(lib): add evaluation count to EA stats

feat: add possibility to evolve strategy during evolution_algorithm

chore: do not pass pairing as mut

tests: tweak one_max parameters

refactor: abstract pairing to n-ary

tests: adjust evolution one_max algorithm to always find optimum

refactor: pass rng as argument

Instead of having the Rng stored
inside the structs, pass it through
the functions. This means it's no longer
necessary to pass perturbations etc. as mutable.

tests: add simple test for evolution on one_max

refactor: do perturbation in place instead of cloning

chore: move env to codes/eoa_lib

feat: add evolution algorithm