allow providing vlm captions with surrounding page text

[edknv]

Description

This PR adds context_text_max_chars parameter to allow enriching VLM image captions with surrounding page text. When enabled, each image's caption prompt is prepended with nearby text, improving retrieval accuracy for documents where images and surrounding text are semantically linked.

Checklist

• I am familiar with the Contributing Guidelines.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.
• If adjusting docker-compose.yaml environment variables have you ensured those are mimicked in the Helm values.yaml file.