Strip markdown decoration from model-emitted insight titles

Models wrap the title line despite the prompt — "**Title: A Day in the
Woods**", "## Title: ...", bold around just the label — which made
parse_title_body's bare "Title:" prefix match fall through to the
fallbacks and leak asterisks into the stored title.

strip_title_markdown trims bold/italic markers, heading hashes,
backticks, and quotes from both ends; applied to the label line, the
extracted title, both fallback paths, and generate_photo_title (which
previously stripped only quotes).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

src/ai/ollama.rs

@@ -548,7 +548,16 @@ Capture the key moment or theme. Return ONLY the title, nothing else."#,
         let title = self
             .generate_with_images(&prompt, Some(system), None)
             .await?;
-        Ok(title.trim().trim_matches('"').to_string())
+        // Models decorate despite "Return ONLY the title": quotes, bold
+        // markers, sometimes a "Title:" label.
+        use crate::ai::insight_generator::strip_title_markdown;
+        let cleaned = strip_title_markdown(title.trim());
+        let cleaned = cleaned
+            .strip_prefix("Title:")
+            .or_else(|| cleaned.strip_prefix("title:"))
+            .map(strip_title_markdown)
+            .unwrap_or(cleaned);
+        Ok(cleaned.to_string())
     }
 
     /// Generate a summary for a single photo based on its context