Support hyper-parameter searching for CF explanation methods

[BirkhoffG]

Supporting hyper-parameter searching enables us to properly benchmark the algorithms. This issue is a thread discussing how to support hyperparameter searching in CF explanation methods.

In essence, this problem is a multi-objective problem (i.e., minimizing invalidity and cost). Bayesian optimization seems to be an obvious choice.

Some open-sourced libraries of hyper-parameter searching:

• Optuna [doc]
• Ax [doc]
• Dragonfly [link]
• Spearmint [link]
• BoTorch [link]

At present, I would prefer to implement our own Bayesian optimization algorithms for hyperparameter searching, instead of relying on these open-source libraries.