[V0 deprecation] Remove _VLLM_V1 suffixes from attention backend names (#25489)

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni001@gmail.com>

vllm/v1/attention/backends/triton_attn.py

@@ -155,7 +155,7 @@ class TritonAttentionBackend(AttentionBackend):
 
     @staticmethod
     def get_name() -> str:
-        return "TRITON_ATTN_VLLM_V1"
+        return "TRITON_ATTN"
 
     @staticmethod
     def get_impl_cls() -> type["TritonAttentionImpl"]:

vllm/v1/attention/backends/xformers.py

@@ -90,7 +90,7 @@ class XFormersAttentionBackend(AttentionBackend):
 
     @staticmethod
     def get_name() -> str:
-        return "XFORMERS_VLLM_V1"
+        return "XFORMERS"
 
     @staticmethod
     def get_impl_cls() -> type["XFormersAttentionImpl"]: