fix(deps): Add transformers[vision] for AutoModelForImageTextToText

Problem: Backend crashes with ModuleNotFoundError: AutoModelForImageTextToText

Solution: Changed transformers (>=4.46.0) to transformers[vision] (>=4.46.0)
to include vision-text model dependencies required by Docling's CodeFormulaModel

Author: manavgup

poetry.lock

@@ -45,7 +45,7 @@ dependencies = [
     "validators>=0.34.0",
     "psutil (>=7.0.0,<8.0.0)",
     "docling (>=2.0.0)",
-    "transformers (>=4.46.0)",
+    "transformers[vision] (>=4.46.0)",
     "pydub (>=0.25.1,<0.26.0)",
     "uuid-extension (>=0.2.0,<0.3.0)",
     "spacy (>=3.7.0,<4.0.0)",