[XPU] move xpu_attn_backend.py to FastDeploy/fastdeploy/model_executor/layers/backends/xpu (#5878)

fastdeploy/model_executor/layers/attention/__init__.py

@@ -23,14 +23,12 @@ from .iluvatar_attn_backend import IluvatarAttnBackend
 from .mla_attention_backend import MLAAttentionBackend
 from .moba_attention_backend import PlasAttentionBackend
 from .native_paddle_backend import PaddleNativeAttnBackend
-from .xpu_attn_backend import XPUAttentionBackend
 
 __all__ = [
     "AttentionBackend",
     "PaddleNativeAttnBackend",
     "get_attention_backend",
     "AppendAttentionBackend",
-    "XPUAttentionBackend",
     "MLAAttentionBackend",
     "FlashAttentionBackend",
     "IluvatarAttnBackend",

fastdeploy/model_executor/layers/backends/xpu/__init__.py

@@ -16,6 +16,7 @@
 xpu backend methods
 """
 
+from .attention import XPUAttentionBackend
 from .moe.fused_moe import XPUMoEMethod, XPUWeightOnlyMoEMethod
 from .quantization.weight_only import XPUWeightOnlyLinearMethod
 
@@ -23,4 +24,5 @@ __all__ = [
     "XPUWeightOnlyLinearMethod",
     "XPUMoEMethod",
     "XPUWeightOnlyMoEMethod",
+    "XPUAttentionBackend",
 ]