Accelerating Dataset Distillation via Model Augmentation