fix(progress): report embedding progress (was showing 100% instantly)

Root cause: Progress was only reported during file extraction (~2s).
Embedding generation takes ~35s with no progress updates.

Fix:
- Added functions_total to progress model
- Percent now weighted: 20% file extraction, 80% embedding
- Reports progress during embedding phase with function count
- UI will now show gradual 0-100% over full indexing duration

Author: DevanshuNEU

backend/routes/repos.py

@@ -292,7 +292,8 @@ async def progress_callback(
                 files_processed: int,
                 functions_found: int,
                 total_files: int,
-                current_file: str = None
+                current_file: str = None,
+                functions_total: int = 0
             ):
                 nonlocal tracked_total_files
                 tracked_total_files = total_files
@@ -301,15 +302,16 @@ async def progress_callback(
                         "Publishing progress event",
                         repo_id=repo_id,
                         files=f"{files_processed}/{total_files}",
-                        functions=functions_found,
+                        functions=f"{functions_found}/{functions_total}" if functions_total else str(functions_found),
                         file=current_file
                     )
                     publisher.publish_progress(
                         repo_id,
                         files_processed,
                         total_files,
                         functions_found,
-                        current_file
+                        current_file,
+                        functions_total
                     )
 
             total_functions = await indexer.index_repository_with_progress(

backend/services/indexer_optimized.py

@@ -768,12 +768,23 @@ async def index_repository_with_progress(
         ]
 
         all_embeddings = []
+        total_to_embed = len(embedding_texts)
         for i in range(0, len(embedding_texts), self.EMBEDDING_BATCH_SIZE):
             batch_texts = embedding_texts[i:i + self.EMBEDDING_BATCH_SIZE]
             batch_embeddings = await self._create_embeddings_batch(batch_texts)
             all_embeddings.extend(batch_embeddings)
 
-            logger.debug("Embeddings generated", completed=len(all_embeddings), total=len(embedding_texts))
+            # Report embedding progress - this is where most time is spent
+            embedded_count = len(all_embeddings)
+            await progress_callback(
+                total_files,  # Files done
+                embedded_count,  # Functions embedded so far
+                total_files,
+                f"Embedding functions ({embedded_count}/{total_to_embed})...",
+                total_to_embed  # Total functions to embed
+            )
+            
+            logger.debug("Embeddings generated", completed=embedded_count, total=total_to_embed)
 
         # Prepare vectors for Pinecone
         logger.debug("Uploading to Pinecone")

backend/services/indexing_events.py

@@ -33,10 +33,18 @@ class IndexingProgress:
     functions_found: int
     current_file: Optional[str] = None
     percent: int = 0
+    functions_total: int = 0  # Total functions to embed (set during embedding phase)
 
     def __post_init__(self):
-        if self.files_total > 0:
-            self.percent = int((self.files_processed / self.files_total) * 100)
+        # If we have functions_total, we're in embedding phase (slow) - weight it 80%
+        # File extraction is fast, weight it 20%
+        if self.functions_total > 0 and self.files_total > 0:
+            file_progress = (self.files_processed / self.files_total) * 20  # 0-20%
+            embed_progress = (self.functions_found / self.functions_total) * 80  # 0-80%
+            self.percent = int(file_progress + embed_progress)
+        elif self.files_total > 0:
+            # Still in file extraction phase (0-20%)
+            self.percent = int((self.files_processed / self.files_total) * 20)
 
 
 @dataclass
@@ -118,14 +126,16 @@ def publish_progress(
         files_processed: int,
         files_total: int,
         functions_found: int,
-        current_file: Optional[str] = None
+        current_file: Optional[str] = None,
+        functions_total: int = 0
     ) -> bool:
         """Publish indexing progress update."""
         progress = IndexingProgress(
             files_processed=files_processed,
             files_total=files_total,
             functions_found=functions_found,
-            current_file=current_file
+            current_file=current_file,
+            functions_total=functions_total
         )
 
         return self._publish(entity_id, {