feat(data-entry): Bulk Receipt Upload cu Mobile UX Android Nativ

## Funcționalități Principale

### Bulk Upload & Processing
- Drag & drop pentru upload bonuri multiple oriunde pe pagină
- Batch processing cu job queue și worker pool
- Real-time updates via SSE (Server-Sent Events) cu fallback polling
- Duplicate detection via SHA-256 file hash
- Auto-retry pentru job-uri failed
- Cancel individual jobs sau batch complet

### Mobile UX - Android Native Style
- Top bar fixă cu hamburger, titlu centrat, acțiuni (search/filter)
- Bottom navigation cu 4 tab-uri (Bonuri, Upload, Rapoarte, Setări)
- FAB (Floating Action Button) cu hide/show on scroll
- Filter chips orizontal scrollabile
- Selecție multiplă prin long-press (500ms)
- Select All + Bulk Delete cu confirmare
- Layout Android pentru Create/Edit/View bon (Gmail compose style)

### Bug Fixes
- Refresh individual via SSE în loc de refresh total pagină
- Bonurile cu eroare OCR rămân vizibile pentru editare manuală
- Afișare nume fișier original pentru toate bonurile
- Upload stabil pe mobil (fix race condition File API)
- Păstrare ordine bonuri la refresh (nu se reordonează)

### Backend
- SSE endpoint pentru status updates real-time
- Bulk delete endpoint cu partial success
- Auto-cleanup bonuri failed după 7 zile
- Batch model cu tracking complet

### Testing
- E2E tests cu Playwright
- Unit tests pentru bulk upload, auto-create, cleanup

## Commits Squashed: 43 user stories (US-001 → US-043)
## Branch: ralph/bulk-receipt-upload
## Timp dezvoltare: ~3 zile (Ralph autonomous)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

backend/modules/data_entry/services/ocr/job_queue.py

@@ -23,7 +23,9 @@ Schema:
         ocr_time_ms INTEGER,         -- Actual OCR engine processing time
         created_by TEXT,           -- Username
         original_filename TEXT,
-        expires_at TIMESTAMP
+        expires_at TIMESTAMP,
+        batch_id INTEGER,          -- Foreign key to batch_uploads (for bulk processing)
+        file_hash TEXT             -- SHA-256 hash for duplicate detection (US-007)
     )
 """
 
@@ -66,6 +68,7 @@ class OCRJobStatus(str, Enum):
     processing = "processing"
     completed = "completed"
     failed = "failed"
+    cancelled = "cancelled"
 
 
 @dataclass
@@ -86,6 +89,8 @@ class OCRJob:
     created_by: Optional[str] = None
     original_filename: Optional[str] = None
     expires_at: Optional[datetime] = None
+    batch_id: Optional[int] = None  # Links to batch_uploads table for bulk processing
+    file_hash: Optional[str] = None  # SHA-256 hash for duplicate detection (US-007)
 
     @property
     def queue_wait_ms(self) -> Optional[int]:
@@ -163,7 +168,8 @@ class OCRJobQueue:
                     ocr_time_ms INTEGER,
                     created_by TEXT,
                     original_filename TEXT,
-                    expires_at TIMESTAMP
+                    expires_at TIMESTAMP,
+                    batch_id INTEGER
                 )
             ''')
 
@@ -174,6 +180,20 @@ class OCRJobQueue:
             except Exception:
                 pass  # Column already exists
 
+            # Migration: add batch_id column if it doesn't exist
+            try:
+                await db.execute('ALTER TABLE ocr_jobs ADD COLUMN batch_id INTEGER')
+                logger.info("[OCRJobQueue] Added batch_id column to existing table")
+            except Exception:
+                pass  # Column already exists
+
+            # Migration: add file_hash column if it doesn't exist (US-007)
+            try:
+                await db.execute('ALTER TABLE ocr_jobs ADD COLUMN file_hash TEXT')
+                logger.info("[OCRJobQueue] Added file_hash column to existing table")
+            except Exception:
+                pass  # Column already exists
+
             # Index for efficient queue queries
             await db.execute('''
                 CREATE INDEX IF NOT EXISTS idx_ocr_jobs_status
@@ -197,7 +217,9 @@ class OCRJobQueue:
         mime_type: str,
         engine: str = "doctr_plus",
         username: Optional[str] = None,
-        original_filename: Optional[str] = None
+        original_filename: Optional[str] = None,
+        batch_id: Optional[int] = None,
+        file_hash: Optional[str] = None
     ) -> OCRJob:
         """
         Create a new OCR job.
@@ -210,6 +232,8 @@ class OCRJobQueue:
             engine: OCR engine ('tesseract', 'doctr', 'doctr_plus', 'paddleocr')
             username: Username of requester
             original_filename: Original filename from upload
+            batch_id: Optional batch ID for bulk upload processing
+            file_hash: Optional SHA-256 hash for duplicate detection (US-007)
 
         Returns:
             Created OCRJob instance
@@ -241,15 +265,15 @@ class OCRJobQueue:
             await db.execute('''
                 INSERT INTO ocr_jobs (
                     id, status, file_path, mime_type, engine,
-                    created_at, created_by, original_filename, expires_at
-                ) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?)
+                    created_at, created_by, original_filename, expires_at, batch_id, file_hash
+                ) VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
             ''', (
                 job_id, OCRJobStatus.pending.value, str(file_path), mime_type, engine,
-                now.isoformat(), username, original_filename, expires_at.isoformat()
+                now.isoformat(), username, original_filename, expires_at.isoformat(), batch_id, file_hash
             ))
             await db.commit()
 
-        logger.info(f"[OCRJobQueue] Created job {job_id}: engine={engine}, file={file_path.name}")
+        logger.info(f"[OCRJobQueue] Created job {job_id}: engine={engine}, file={file_path.name}, batch_id={batch_id}")
 
         return OCRJob(
             id=job_id,
@@ -260,7 +284,9 @@ class OCRJobQueue:
             created_at=now,
             created_by=username,
             original_filename=original_filename,
-            expires_at=expires_at
+            expires_at=expires_at,
+            batch_id=batch_id,
+            file_hash=file_hash
         )
 
     async def get_job(self, job_id: str) -> Optional[OCRJob]:
@@ -601,6 +627,8 @@ class OCRJobQueue:
             created_by=row['created_by'],
             original_filename=row['original_filename'],
             expires_at=parse_datetime(row['expires_at']),
+            batch_id=row['batch_id'] if 'batch_id' in row.keys() else None,
+            file_hash=row['file_hash'] if 'file_hash' in row.keys() else None,
         )