Model Attribution in Machine-Generated Disinformation: A Domain Generalization Approach with Supervised Contrastive Learning