Modal boosts SGLang multimodal inference over 10% with a Python dict cache

Modal profiled SGLang's scheduler on a multimodal workload and replaced repeated CUDA IPC handle book-keeping with a simple Python dict cache, lifting throughput 16.2% and cutting mean TTFT 13.2% and TPOT 17.2% on Qwen2.5-VL-3B-Instruct on H100s. The fix is merged in SGLang v0.5.10.