LDSDet: Long-Range Context and Dynamic Cross-Modal Alignment for Multimodal Object Detection Under Challenging Illumination | Synapse