MVFusion: Generative Representation Learning With Masked Variational Autoencoders for Multi-Modality Image Fusion. — SciRadar