Fix set_paddle_mkldnn python interface (#328)

* fd serving add dockerfile

* fix enable_paddle_mkldnn

* delete disable_paddle_mkldnn

* fix python set_paddle_mkldnn

Co-authored-by: Jason <jiangjiajun@baidu.com>

docs/api/runtime/runtime_option.md

@@ -66,7 +66,7 @@ use_openvino_backend()
 使用OpenVINO后端进行推理，支持CPU, 支持Paddle/ONNX模型格式
 
 ```
-set_paddle_mkldnn()
+set_paddle_mkldnn(pd_mkldnn=True)
 ```
 当使用Paddle Inference后端时，通过此开关开启或关闭CPU上MKLDNN推理加速，后端默认为开启
 

docs/docs_en/api/runtime/runtime_option.md

@@ -73,7 +73,7 @@ use_openvino_backend()
 Inference with OpenVINO backend (CPU supported, Paddle/ONNX model format supported)
 
 ```
-set_paddle_mkldnn()
+set_paddle_mkldnn(pd_mkldnn=True)
 ```
 
 When using the Paddle Inference backend, this parameter determines whether the MKLDNN inference acceleration on the CPU is on or off. It is on by default.

python/fastdeploy/runtime.py

@@ -85,8 +85,8 @@ class RuntimeOption:
     def use_lite_backend(self):
         return self._option.use_lite_backend()
 
-    def set_paddle_mkldnn(self):
-        return self._option.set_paddle_mkldnn()
+    def set_paddle_mkldnn(self, pd_mkldnn=True):
+        return self._option.set_paddle_mkldnn(pd_mkldnn)
 
     def enable_paddle_log_info(self):
         return self._option.enable_paddle_log_info()