feat: implement hybrid image extraction and memory management

Backend:
- Add hybrid image extraction for Direct track (inline image blocks)
- Add render_inline_image_regions() fallback when OCR doesn't find images
- Add check_document_for_missing_images() for detecting missing images
- Add memory management system (MemoryGuard, ModelManager, ServicePool)
- Update pdf_generator_service to handle HYBRID processing track
- Add ElementType.LOGO for logo extraction

Frontend:
- Fix PDF viewer re-rendering issues with memoization
- Add TaskNotFound component and useTaskValidation hook
- Disable StrictMode due to react-pdf incompatibility
- Fix task detail and results page loading states

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

backend/app/models/unified_document.py

@@ -212,26 +212,44 @@ class TableData:
         if self.caption:
             html.append(f"<caption>{self.caption}</caption>")
 
-        # Group cells by row
-        rows_data = {}
+        # Group cells by row and column for quick lookup
+        cell_map = {}
         for cell in self.cells:
-            if cell.row not in rows_data:
-                rows_data[cell.row] = []
-            rows_data[cell.row].append(cell)
+            cell_map[(cell.row, cell.col)] = cell
 
-        # Generate HTML
+        # Track which cells are covered by row/col spans
+        covered = set()
+        for cell in self.cells:
+            if cell.row_span > 1 or cell.col_span > 1:
+                for r in range(cell.row, cell.row + cell.row_span):
+                    for c in range(cell.col, cell.col + cell.col_span):
+                        if (r, c) != (cell.row, cell.col):
+                            covered.add((r, c))
+
+        # Generate HTML with proper column filling
         for row_idx in range(self.rows):
             html.append("<tr>")
-            if row_idx in rows_data:
-                for cell in sorted(rows_data[row_idx], key=lambda c: c.col):
+            for col_idx in range(self.cols):
+                # Skip cells covered by row/col spans
+                if (row_idx, col_idx) in covered:
+                    continue
+
+                cell = cell_map.get((row_idx, col_idx))
+                tag = "th" if row_idx == 0 and self.headers else "td"
+
+                if cell:
                     span_attrs = []
                     if cell.row_span > 1:
                         span_attrs.append(f'rowspan="{cell.row_span}"')
                     if cell.col_span > 1:
                         span_attrs.append(f'colspan="{cell.col_span}"')
                     span_str = " ".join(span_attrs)
-                    tag = "th" if row_idx == 0 and self.headers else "td"
-                    html.append(f'<{tag} {span_str}>{cell.content}</{tag}>')
+                    content = cell.content if cell.content else ""
+                    html.append(f'<{tag} {span_str}>{content}</{tag}>')
+                else:
+                    # Fill in empty cell for missing positions
+                    html.append(f'<{tag}></{tag}>')
+
             html.append("</tr>")
 
         html.append("</table>")