refactor(tsp): Do not allocate in binary string fitness function

Currently the fit function has allocated for each individual!
That slows it down terribly!

This reimplements it to implement the fit_population, reusing
the same node permutation and same precedence_count

chore: fix warnings

feat(lib): allow modifying fitness function in evolution_algorithm

chore: move population structs to separate module

feat(tsp): Remove with iteration functions

feat: split params out of PerturbationOperator to allow for non-'static PerturbationOperator

This means PerturbationOperators can now hold non-'static references.
Right away, utilize this in the Random2OptPerturbation for the TSPInstance.

feat(tsp): final tweaks

fix(tsp): few more tweaks

chore: add jsons with plot definitions

feat(tsp): add binary ls, use 10k iterations instead of 5k

feat: add a lot of algorithm showcases

chore(tsp): few minor updates

tests(tsp): move tests to proper modules, fix uses

feat(tsp): add MinimalSpanningTree initializer

feat(tsp): add 2opt perturbation for subsequence reversal

feat(tsp): add TSPInstance convertor to GenericGraph

feat(tsp): add minimal spanning tree and filter_edges

feat(tsp): add nearest neighbor initializer

feat(tsp): add cycle, no and partially mapped crossovers

refactor(tsp): split to multiple files out of tsp.rs