Fix image search: embed markdown instead of describing URLs

- tools.py: search_images result now includes 'embed' (ready-to-use
  markdown image syntax) and 'citation' fields instead of raw 'local_url';
  adds 'instructions' field so the model knows to render them verbatim
- llm.py: system prompt now explicitly tells the model to embed images
  using the 'embed' field rather than describing or listing URLs
- markdown.ts: explicitly allow src/alt in PURIFY_OPTS_FULL so img tags
  are never stripped by DOMPurify

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

src/fabledassistant/services/llm.py

@@ -398,6 +398,11 @@ async def build_context(
         tool_lines.append("Use search_todos to find a specific CalDAV todo by keyword when list_todos would return too many results.")
     tool_lines.append("For relative dates like 'Friday' or 'next week', resolve them to YYYY-MM-DD format.")
     tool_lines.append("When creating notes, use the `tags` parameter — do not embed #tag text in the note body.")
+    tool_lines.append(
+        "When search_images returns results, embed each image directly in your response by writing "
+        "the 'embed' field verbatim (e.g. ![Cat](/api/images/3)), then the 'citation' field on the "
+        "next line. Never describe images as text or list their URLs — always render them as markdown images."
+    )
     tool_lines.append(
         "Use update_note to edit/expand an existing note OR to update a task's status/priority/due_date. "
         "Use create_note ONLY for genuinely new notes with a different title. "