fix en_PP-OCR visua (PaddlePaddle#4355)

Author: zhangyubo0722

paddlex/inference/models/text_recognition/predictor.py

@@ -111,7 +111,7 @@ def foo(self, *args, **kwargs):
         return None, None
 
     def get_vis_font(self):
-        if self.model_name.startswith("PP-OCR"):
+        if self.model_name.startswith(("PP-OCR", "en_PP-OCR")):
             return SIMFANG_FONT
 
         if self.model_name in (