Implement "Sampling" fairness-aware preprocessing

[cosmicBboy]

Sampling is composed of two methods:

Uniform Sampling

• uniformly sample (with replacement) n observations from each group, where n is the expected size of that group assuming a uniform distribution (conditioned on the protected class s).

Preferential Sampling

• sample observations using a ranker R, similar to the massaging method.
• the procedure is to duplicate the top-ranked X_s1_y+ and X_s0_y– while removing
  top-ranked X_s1_y– and X_s0_y+.