feat(ai): hybrid backend mode for agentic insights

Adds a `backend` column to photo_insights (default 'local', migration
2026-04-20-000000) and a corresponding optional `backend` field on the
agentic request. When a request sets backend=hybrid:

- The local Ollama vision model is called once via describe_image to
  produce a text description.
- The description is inlined into the first user message as text —
  no base64 image is ever sent to the chat model.
- The agentic tool-calling loop and title generation route through an
  OpenRouterClient (dispatched via &dyn LlmClient), letting the user
  pick any tool-capable model from OpenRouter per request.
- describe_photo is removed from the offered tools since the description
  is already present.

Embeddings and vision stay on local Ollama regardless of backend.
Hybrid mode requires OPENROUTER_API_KEY; handlers return a clear error
when hybrid is requested without it, and also when the selected
OpenRouter model lacks tool-calling support.

AppState gains an optional openrouter client built from
OPENROUTER_API_KEY / OPENROUTER_BASE_URL / OPENROUTER_DEFAULT_MODEL /
OPENROUTER_EMBEDDING_MODEL / attribution headers. Default model is
anthropic/claude-sonnet-4.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/ai/handlers.rs

@@ -28,6 +28,10 @@ pub struct GeneratePhotoInsightRequest {
     pub top_k: Option<i32>,
     #[serde(default)]
     pub min_p: Option<f32>,
+    /// `"local"` (default, Ollama with images) | `"hybrid"` (local vision +
+    /// OpenRouter chat). Only respected by the agentic endpoint.
+    #[serde(default)]
+    pub backend: Option<String>,
 }
 
 #[derive(Debug, Deserialize)]
@@ -65,6 +69,7 @@ pub struct PhotoInsightResponse {
     pub eval_count: Option<i32>,
     #[serde(skip_serializing_if = "Option::is_none")]
     pub approved: Option<bool>,
+    pub backend: String,
 }
 
 #[derive(Debug, Serialize)]
@@ -187,6 +192,7 @@ pub async fn get_insight_handler(
                 prompt_eval_count: None,
                 eval_count: None,
                 approved: insight.approved,
+                backend: insight.backend,
             };
             HttpResponse::Ok().json(response)
         }
@@ -254,6 +260,7 @@ pub async fn get_all_insights_handler(
                     prompt_eval_count: None,
                     eval_count: None,
                     approved: insight.approved,
+                    backend: insight.backend,
                 })
                 .collect();
 
@@ -309,6 +316,10 @@ pub async fn generate_agentic_insight_handler(
         max_iterations
     );
 
+    if let Some(ref b) = request.backend {
+        span.set_attribute(KeyValue::new("backend", b.clone()));
+    }
+
     let result = insight_generator
         .generate_agentic_insight_for_photo(
             &normalized_path,
@@ -320,6 +331,7 @@ pub async fn generate_agentic_insight_handler(
             request.top_k,
             request.min_p,
             max_iterations,
+            request.backend.clone(),
         )
         .await;
 
@@ -341,6 +353,7 @@ pub async fn generate_agentic_insight_handler(
                         prompt_eval_count,
                         eval_count,
                         approved: insight.approved,
+                        backend: insight.backend,
                     };
                     HttpResponse::Ok().json(response)
                 }