1.1.0

• fixed outcome_levels issue when levels of provided predictions do not match outcome levels
• renamed outcome_levels to preds_levels to improve clarity
• added outcome_base argument to set base level for target variable used to compute fairness metrics
• fixed fnr_parity() and fpr_parity() calculations for different outcome bases
• updates in package documentation