Extract all text elements from this document image with detailed formatting information.

For each text element, provide:
- bbox_2d: [x1, y1, x2, y2] bounding box coordinates
- text_content: The text content (merge consecutive lines of same paragraph with space)
- element_type: "title", "header", "paragraph", "label", "value", "table_cell", "list_item", "footer"
- style:
  - bold: true/false
  - italic: true/false
  - font_size: "small", "normal", "large", "xlarge"
  - alignment: "left", "center", "right"

Important:
- MERGE consecutive lines that form a single paragraph into ONE element
- Keep separate elements for: titles, headers, labels, values, footers
- For label:value pairs, keep them as separate elements

Output format: JSON array