[misc] Add FlashMLA as a new option of VLLM_ATTENTION_BACKEND env (#14267)

vllm/envs.py

@@ -321,6 +321,7 @@ environment_variables: dict[str, Callable[[], Any]] = {
     # - "XFORMERS": use XFormers
     # - "ROCM_FLASH": use ROCmFlashAttention
     # - "FLASHINFER": use flashinfer
+    # - "FLASHMLA": use FlashMLA
     "VLLM_ATTENTION_BACKEND":
     lambda: os.getenv("VLLM_ATTENTION_BACKEND", None),