fix: torch dependency start error (infiniflow#2777)

### What problem does this PR solve?

when use slim image, remove ```torch``` denpendency.

### Type of change

- [✓] Bug Fix (non-breaking change which fixes an issue)

---------

Co-authored-by: chongchuanbing <chongchuanbing@gmail.com>
Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>

deepdoc/parser/pdf_parser.py

@@ -45,9 +45,12 @@ def __init__(self):
 
         self.updown_cnt_mdl = xgb.Booster()
         if not LIGHTEN:
-            import torch
-            if torch.cuda.is_available():
-                self.updown_cnt_mdl.set_param({"device": "cuda"})
+            try:
+                import torch
+                if torch.cuda.is_available():
+                    self.updown_cnt_mdl.set_param({"device": "cuda"})
+            except Exception as e:
+                logging.error(str(e))
         try:
             model_dir = os.path.join(
                 get_project_base_directory(),