[TabArena-v0.X] Tune criterion for RF / XT

For classification with RF/XT, we could optimize the criterion in `['gini', 'entropy']`. In my tests, one could just always use 'entropy' if logloss is the target metric, but I suspect that 'gini' might be better for other target metrics.
Since the criterion needs to be different for classification and regression, one could add a hyperparameter `classification_criterion` which is tuned in `['gini', 'entropy']` and ignored for regression. This might require changing AutoGluon code, though.

Start by tracing how RF/XT classification and regression settings are represented in TabArena and where AutoGluon configuration is passed. Check how target metrics are selected, then verify that classification can tune `classification_criterion` over `['gini', 'entropy']` while regression ignores it. Done means the benchmark runs both classification criteria and preserves separate regression behavior.

Written by the indexing model from the issue text.