Feature Request: Capture and Reconstruct Images in Google Docs

[saminur]

Currently, WritingObserver does not capture or reconstruct images inserted into Google Docs (e.g., images uploaded from a local device or inserted via the toolbar). This results in incomplete document reconstruction and loss of important visual context.

Expected Behavior:

• Detect when an image is inserted into a Google Doc.
• Capture relevant metadata, such as: 1) Document ID 2) Tab/section ID 3) Approximate position (relative to surrounding text or structural elements)
• Store a reference (URL or Docs object ID) so the image can be reconstructed or represented during replay.