Implement Parameter search

[jrcalgo]

Built a parameter search script that utilizes symbolic regression to aid explainability and bayesian optimization on said regressor to optimize the selection of hyperparameters.

This script uses data from config.json's algorithm hyperparams and training data's performance results. Performance result data can either encompass a singular training instance or an entire algorithm's training directory for a given example environment.

[jrcalgo]

Besides the search I have already written on this branch, a suite of param searchers will be very useful for minimizing hyperparameter adjustment-induced headaches