A Large‐Scale Empirical Evaluation of Cross‐Validation and External Test Set Validation in (Q)SAR | Synapse