Align `eval_set` with scikit-learn

[jinlow]

Consider accepting eval_set and sample_weight_eval_set in the same way XGBoost scikit-learn does. This will just make the API easier to use directly in XGBoost scikit-learn pipelines.