[Rebase] Rebase to vllm v0.16.0 (#1357)

Signed-off-by: Taichang Zhou <tzhouam@connect.ust.hk>
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
Co-authored-by: 汪志鹏 <wangzhipeng628@gmail.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Co-authored-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: ZJY0516 <zhu.jiangyun@foxmail.com>
Co-authored-by: youkaichao <youkaichao@gmail.com>

requirements/cuda.txt

@@ -1,3 +1,5 @@
 -r common.txt
 onnxruntime>=1.23.2
-fa3-fwd==0.0.1
+fa3-fwd==0.0.2
+# nvidia-cublas-cu12 is pinned to 12.9.1.4 via force-reinstall in Dockerfile.ci
+# because torch pins its own cublas version, making it unsatisfiable at resolve time.