[MetaxGPU] Support FastDeploy on metax gpu (#3241)

* [MetaxGPU] Support FastDeploy on metax gpu

* Update metax_worker.py

1. change worker log;
2. remove custom allreduce, adapt it later;
3. remove cuda graph;

* Update __init__.py

1. remove metax's key work comment

* Update __init__.py

1. remove metax's key word comment;
2. add fused_moe_kernel_paddle import

---------

Co-authored-by: yongqiangma <xing.wo@163.com>

fastdeploy/model_executor/layers/backends/__init__.py

@@ -48,3 +48,10 @@ if current_platform.is_dcu():
     if hasattr(dcu, "__all__"):
         globals().update({name: getattr(dcu, name) for name in dcu.__all__})
         __all__.extend(dcu.__all__)
+
+if current_platform.is_maca():
+    from . import metax
+
+    if hasattr(metax, "__all__"):
+        globals().update({name: getattr(metax, name) for name in metax.__all__})
+        __all__.extend(metax.__all__)