fix(ocr): sync workflow steps with process_with_ocr replacing legacy azure step

- Update _compute_processing_flow to recognize process_with_ocr as the OCR
  stage and remap legacy process_with_azure_document_intelligence log entries
  for backward compatibility
- Normalize legacy OCR step name in _compute_step_summary log fallback
- Add process_with_ocr to REAL_MAIN_STEPS/REAL_STEPS in step_manager,
  file_status, and file_queries (keeping legacy name for old DB entries)
- Update retry logic in api/files.py to retry failed OCR via process_with_ocr
  (handles both step names as aliases)
- Fix process_document.py to log process_with_ocr as skipped (not azure step)
  for the local text extraction path

Co-authored-by: christianlouis <361235+christianlouis@users.noreply.github.com>

app/utils/file_queries.py

@@ -21,9 +21,9 @@ def apply_status_filter(query: Query, db: Session, status: Optional[str]) -> Que
     processing status by examining associated FileProcessingStep entries.
 
     Only tracks "real" processing steps that represent user-facing status:
-    - Main steps: create_file_record, check_text, extract_text, process_with_azure_document_intelligence,
-                  extract_metadata_with_gpt, embed_metadata_into_pdf, finalize_document_storage,
-                  send_to_all_destinations
+    - Main steps: create_file_record, check_text, extract_text, process_with_ocr,
+                  process_with_azure_document_intelligence (legacy), extract_metadata_with_gpt,
+                  embed_metadata_into_pdf, finalize_document_storage, send_to_all_destinations
     - Upload steps: queue_*, upload_to_*
 
     Diagnostic/internal steps (poll_task, upload_file, set_custom_fields, etc.) are ignored
@@ -58,6 +58,7 @@ def apply_status_filter(query: Query, db: Session, status: Optional[str]) -> Que
         "create_file_record",
         "check_text",
         "extract_text",
+        "process_with_ocr",
         "process_with_azure_document_intelligence",
         "extract_metadata_with_gpt",
         "embed_metadata_into_pdf",

app/utils/file_status.py

@@ -56,9 +56,9 @@ def get_files_processing_status(db: Session, file_ids: List[int]) -> Dict[int, D
     Get processing status for multiple files efficiently.
 
     Only counts "real" processing steps that represent user-facing status:
-    - Main steps: create_file_record, check_text, extract_text, process_with_azure_document_intelligence,
-                  extract_metadata_with_gpt, embed_metadata_into_pdf, finalize_document_storage,
-                  send_to_all_destinations
+    - Main steps: create_file_record, check_text, extract_text, process_with_ocr,
+                  process_with_azure_document_intelligence (legacy), extract_metadata_with_gpt,
+                  embed_metadata_into_pdf, finalize_document_storage, send_to_all_destinations
     - Upload steps: upload_to_*
 
     Diagnostic/internal steps (poll_task, upload_file, set_custom_fields, etc.) are ignored.
@@ -83,6 +83,7 @@ def get_files_processing_status(db: Session, file_ids: List[int]) -> Dict[int, D
         "create_file_record",
         "check_text",
         "extract_text",
+        "process_with_ocr",
         "process_with_azure_document_intelligence",
         "extract_metadata_with_gpt",
         "embed_metadata_into_pdf",

app/utils/step_manager.py

@@ -201,10 +201,13 @@ def get_file_overall_status(db: Session, file_id: int) -> Dict:
 
     # Define which steps are "real" status-determining steps
     # Only high-level logical steps, not implementation sub-steps
+    # Both process_with_ocr (current) and process_with_azure_document_intelligence (legacy)
+    # are included to correctly count steps for files processed before the OCR abstraction.
     REAL_MAIN_STEPS = {
         "create_file_record",
         "check_text",
         "extract_text",
+        "process_with_ocr",
         "process_with_azure_document_intelligence",
         "extract_metadata_with_gpt",
         "embed_metadata_into_pdf",
@@ -286,10 +289,13 @@ def get_step_summary(db: Session, file_id: int) -> Dict:
     """
     # Define which steps are "real" status-determining steps
     # Only high-level logical steps, not implementation sub-steps
+    # Both process_with_ocr (current) and process_with_azure_document_intelligence (legacy)
+    # are included to correctly count steps for files processed before the OCR abstraction.
     REAL_MAIN_STEPS = {
         "create_file_record",
         "check_text",
         "extract_text",
+        "process_with_ocr",
         "process_with_azure_document_intelligence",
         "extract_metadata_with_gpt",
         "embed_metadata_into_pdf",