fix: change default OCR language from English to Swedish

Project targets Swedish invoice extraction. PaddleOCR sv model provides
better recognition of Swedish-specific characters (å, ä, ö).

packages/backend/backend/cli/infer.py

@@ -54,8 +54,8 @@ def main():
     )
     parser.add_argument(
         '--lang',
-        default='en',
-        help='OCR language (default: en)'
+        default='sv',
+        help='OCR language (default: sv)'
     )
     parser.add_argument(
         '--gpu',

packages/backend/backend/pipeline/field_extractor.py

@@ -85,7 +85,7 @@ class FieldExtractor:
 
     def __init__(
         self,
-        ocr_lang: str = 'en',
+        ocr_lang: str = 'sv',
         use_gpu: bool = False,
         bbox_padding: float = 0.1,
         dpi: int = 300,

packages/backend/backend/pipeline/pipeline.py

@@ -209,7 +209,7 @@ class InferencePipeline:
         self,
         model_path: str | Path,
         confidence_threshold: float = 0.5,
-        ocr_lang: str = 'en',
+        ocr_lang: str = 'sv',
         use_gpu: bool = False,
         dpi: int = 300,
         enable_fallback: bool = True,

packages/backend/backend/web/services/autolabel.py

@@ -37,7 +37,7 @@ class AutoLabelService:
     def ocr_engine(self) -> OCREngine:
         """Lazy initialization of OCR engine."""
         if self._ocr_engine is None:
-            self._ocr_engine = OCREngine(lang="en")
+            self._ocr_engine = OCREngine(lang="sv")
         return self._ocr_engine
 
     def auto_label_document(