feat: detailed embed progress — current section, global bar, queue display

- Backend: tracks current_table, current_inserted, current_total, queue per embed task
- Frontend: dual progress bars (section + global), queue of upcoming sections
- Section bar shows table name + individual progress
- Global bar shows inserted/skipped/total with green/red segments
- Queue shows remaining sections with doc counts

backend/app.py

@@ -3885,36 +3885,44 @@ async def embed_report(rid: str, req: dict, bg: BackgroundTasks, u=Depends(requi
         "message": f"Embedding {total_docs} documentos em {len(tables_used)} tabelas..."
     })
 
+    # Build queue info: [(table, doc_count), ...]
+    table_queue = [(tbl, len(docs)) for tbl, docs in table_docs.items()]
+
     def _bg_embed_all():
         """Background: embed documents into their respective tables sequentially."""
         import array
         default_model = cfg.get("embedding_model_id", "cohere.embed-v4.0")
         inserted_total = 0
+        skipped_total = 0
+        processed_total = 0
         errors = []
-        for tbl, docs in table_docs.items():
+        for tbl_idx, (tbl, docs) in enumerate(table_docs.items()):
+            remaining = table_queue[tbl_idx + 1:] if tbl_idx + 1 < len(table_queue) else []
             _auto_register_table(cfg["id"], tbl)
-            # Purge old data for this tenancy/date before inserting
             purged = _purge_table_by_tenancy(cfg, tbl, tenancy, extract_date)
             if purged:
-                _update_embed_status(task_id, {"message": f"Purged {purged} old docs from {tbl}..."})
-            # Auto-detect embedding model based on table dimension
+                _update_embed_status(task_id, {"message": f"Purged {purged} old docs from {tbl}...",
+                                               "current_table": tbl, "current_inserted": 0, "current_total": len(docs),
+                                               "queue": [{"table": t, "docs": n} for t, n in remaining]})
             emb_model = default_model
             try:
                 actual_dim = _get_table_embedding_dim(cfg, tbl)
                 if actual_dim and actual_dim in _DIM_TO_MODEL:
                     emb_model = _DIM_TO_MODEL[actual_dim]
-                    log.info(f"Table {tbl}: dim={actual_dim}, using model {emb_model}")
-            except Exception as e:
-                log.warning(f"Could not detect dim for {tbl}: {e}")
+            except Exception:
+                pass
+            current_inserted = 0
+            current_skipped = 0
             try:
                 conn = _get_adb_connection(cfg)
                 cur = conn.cursor()
                 for doc in docs:
-                    processed = 0
                     try:
                         content = doc.get("content", "")
                         if not content:
-                            processed += 1
+                            skipped_total += 1
+                            current_skipped += 1
+                            processed_total += 1
                             continue
                         embedding = _embed_text(content, gc, emb_model)
                         vec = array.array('f', [float(x) for x in embedding])
@@ -3926,17 +3934,29 @@ async def embed_report(rid: str, req: dict, bg: BackgroundTasks, u=Depends(requi
                         cur.execute(f'INSERT INTO "{tbl}" (ID, TEXT, EMBEDDING, METADATA) VALUES (HEXTORAW(:1), :2, :3, :4)',
                                     [uuid.uuid4().hex.upper(), content, vec, metadata])
                         inserted_total += 1
+                        current_inserted += 1
                     except Exception as e:
-                        log.warning(f"Embed skip in {tbl}: {str(e)[:120]}")
-                    processed += 1
+                        err_str = str(e)
+                        if "message" in err_str:
+                            import re as _re
+                            m = _re.search(r'"message"\s*:\s*"(.*?)"', err_str)
+                            log.warning(f"Embed skip in {tbl}: {m.group(1)[:200] if m else err_str[:200]}")
+                        else:
+                            log.warning(f"Embed skip in {tbl}: {err_str[:200]}")
+                        skipped_total += 1
+                        current_skipped += 1
+                    processed_total += 1
                     _update_embed_status(task_id, {
-                        "inserted": inserted_total,
-                        "message": f"Embedding {inserted_total}/{processed} OK — {tbl}"
+                        "inserted": inserted_total, "skipped": skipped_total, "processed": processed_total,
+                        "current_table": tbl, "current_inserted": current_inserted, "current_skipped": current_skipped,
+                        "current_total": len(docs),
+                        "queue": [{"table": t, "docs": n} for t, n in remaining],
+                        "message": f"{tbl}: {current_inserted}/{len(docs)} — global: {inserted_total}/{total_docs}"
                     })
                 conn.commit()
                 cur.close()
                 conn.close()
-                log.info(f"Embedded {len(docs)} docs into {tbl} (tenancy={tenancy})")
+                log.info(f"Embedded {current_inserted}/{len(docs)} docs into {tbl} (tenancy={tenancy})")
             except Exception as e:
                 log.error(f"Failed to connect/embed to {tbl}: {e}")
                 errors.append(f"{tbl}: {str(e)[:100]}")