[Model] Gemma4: robust quantized MoE expert weight loading#39406

yueshen2016 wants to merge 1 commit intovllm-project:mainvllm-project/vllm:mainfrom

yueshen2016:fix/gemma4-quantized-moe-loadingyueshen2016/vllm:fix/gemma4-quantized-moe-loadingCopy head branch name to clipboard