chinese_text_detection_db_server模型库，load_inference_model和save_inference_model参数问题

[Zrincet]

https://github.com/PaddlePaddle/PaddleOCR/issues/7456 请提供下述完整信息以便快速定位问题/Please provide the following information to quickly locate the problem

• 系统环境/System Environment：Windows11
• 版本号/Version： Paddle： 2.3.2
• 问题相关组件/Related components：chinese_text_detection_db_server 1.0.3 chinese_ocr_db_crnn_server 1.1.3
• 运行指令/Command Code：

import paddlehub as hub
import cv2

ocr = hub.Module(name="chinese_ocr_db_crnn_server", enable_mkldnn=True)       # mkldnn加速仅在CPU下有效

ocr.save_inference_model(dirname="model/mobilenet")

• 完整报错/Complete Error Message：

发生异常: TypeError
missing a required argument: 'path_prefix'

During handling of the above exception, another exception occurred:

  File "C:\Users\xxx\.paddlehub\modules\chinese_text_detection_db_server\module.py", line 242, in save_inference_model
    program, feeded_var_names, target_vars = paddle.static.load_inference_model(dirname=self.pretrained_model_path,
  File "C:\Users\xxx\.paddlehub\modules\chinese_ocr_db_crnn_server\module.py", line 422, in _save_detector_model
    self.text_detector_module.save_inference_model(dirname, model_filename, params_filename, combined)
  File "C:\Users\xxx\.paddlehub\modules\chinese_ocr_db_crnn_server\module.py", line 416, in save_inference_model
    self._save_detector_model(detector_dir, model_filename, params_filename, combined)
  File "D:\WorkSpace\Python\console-vscode\main.py", line 10, in <module>
    ocr.save_inference_model(dirname="model/mobilenet")

\chinese_text_detection_db_server\module.py中的save_inference_model方法中调用的paddle.static.load_inference_model和paddle.static.save_inference_model的入参与PaddlePaddle官网API文档参数不一致，导致无法保存模型

[jm12138]

目前导出模型接口存在一些兼容性问题，可以尝试使用旧版本进行导出

hub install chinese_text_detection_db_server==1.0.2
hub install chinese_ocr_db_crnn_server==1.1.2