feat: migrate VIN/receipt extractors and OCR service to engine abstraction (refs #117)

Replace direct pytesseract calls with OcrEngine interface in vin_extractor.py,
receipt_extractor.py, and ocr_service.py. PSM mode fallbacks replaced with
engine-agnostic single-line/single-word configs. Dead _process_ocr_data removed.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

ocr/app/extractors/receipt_extractor.py

@@ -1,16 +1,13 @@
 """Receipt-specific OCR extractor with field extraction."""
-import io
 import logging
 import time
 from dataclasses import dataclass, field
 from typing import Any, Optional
 
 import magic
-import pytesseract
-from PIL import Image
 from pillow_heif import register_heif_opener
 
-from app.config import settings
+from app.engines import OcrConfig, create_engine
 from app.extractors.base import BaseExtractor
 from app.preprocessors.receipt_preprocessor import receipt_preprocessor
 from app.patterns import currency_matcher, date_matcher, fuel_matcher
@@ -53,8 +50,8 @@ class ReceiptExtractor(BaseExtractor):
     }
 
     def __init__(self) -> None:
-        """Initialize receipt extractor."""
-        pytesseract.pytesseract.tesseract_cmd = settings.tesseract_cmd
+        """Initialize receipt extractor with engine from factory."""
+        self._engine = create_engine()
 
     def extract(
         self,
@@ -150,26 +147,19 @@ class ReceiptExtractor(BaseExtractor):
         detected = mime.from_buffer(file_bytes)
         return detected or "application/octet-stream"
 
-    def _perform_ocr(self, image_bytes: bytes, psm: int = 6) -> str:
+    def _perform_ocr(self, image_bytes: bytes) -> str:
         """
-        Perform OCR on preprocessed image.
+        Perform OCR on preprocessed image via engine abstraction.
 
         Args:
             image_bytes: Preprocessed image bytes
-            psm: Tesseract page segmentation mode
-                 4 = Assume single column of text
-                 6 = Uniform block of text
 
         Returns:
             Raw OCR text
         """
-        image = Image.open(io.BytesIO(image_bytes))
-
-        # Configure Tesseract for receipt OCR
-        # PSM 4 works well for columnar receipt text
-        config = f"--psm {psm}"
-
-        return pytesseract.image_to_string(image, config=config)
+        config = OcrConfig()
+        result = self._engine.recognize(image_bytes, config)
+        return result.text
 
     def _detect_receipt_type(self, text: str) -> str:
         """