Fix: duplicate output by async_chat_streamly (#11842)

### What problem does this PR solve?

Fix: duplicate output by async_chat_streamly
Refact: revert manual modification

### Type of change

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

---------

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

rag/app/manual.py

@@ -219,15 +219,13 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
         )
 
         def _normalize_section(section):
-            # Pad/normalize to (txt, layout, positions)
-            if not isinstance(section, (list, tuple)):
-                section = (section, "", [])
-            elif len(section) == 1:
+            # pad section to length 3: (txt, sec_id, poss)
+            if len(section) == 1:
                 section = (section[0], "", [])
             elif len(section) == 2:
                 section = (section[0], "", section[1])
-            else:
-                section = (section[0], section[1], section[2])
+            elif len(section) != 3:
+                raise ValueError(f"Unexpected section length: {len(section)} (value={section!r})")
 
             txt, layoutno, poss = section
             if isinstance(poss, str):
@@ -238,12 +236,9 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
                     if isinstance(pn, list) and pn:
                         pn = pn[0]           # [pn] -> pn
                         poss[0] = (pn, *first[1:])
-            if not poss:
-                poss = []
 
             return (txt, layoutno, poss)
 
-
         sections = [_normalize_section(sec) for sec in sections]
 
         if not sections and not tbls:

rag/llm/chat_model.py

@@ -146,7 +146,6 @@ class Base(ABC):
             request_kwargs["stop"] = stop
 
         response = await self.async_client.chat.completions.create(**request_kwargs)
-
         async for resp in response:
             if not resp.choices:
                 continue
@@ -161,7 +160,6 @@ class Base(ABC):
             else:
                 reasoning_start = False
                 ans = resp.choices[0].delta.content
-
             tol = total_token_count_from_response(resp)
             if not tol:
                 tol = num_tokens_from_string(resp.choices[0].delta.content)