feat: add Vision LLM integration (CLIP + Qwen3-VL cascade)

- Add Qwen3-VL dynamic management (start/stop/status CLI)
- Add CLIP + Qwen3-VL cascade detection strategy
- Add Vision CLI commands (vision start/stop/status, detect)
- Add cascade_vision processor module
- Add clip processor module
- Add qwen_vl_manager module

Changes:
- scripts/start_qwen3vl.sh, stop_qwen3vl.sh: Qwen3-VL management scripts
- src/core/vision/: Qwen3-VL manager module
- src/core/processor/cascade_vision.rs: CLIP + Qwen3-VL cascade logic
- src/core/processor/clip.rs: CLIP classification and detection
- src/api/clip_api.rs: CLIP API endpoints
- src/cli/vision.rs: Vision CLI implementation
- src/cli/args.rs: Add Vision and Detect commands
- src/main.rs: Integrate Vision CLI
- src/core/mod.rs: Add vision module
- src/core/processor/mod.rs: Add cascade_vision module

src/api/identity_binding.rs

@@ -734,6 +734,8 @@ pub async fn bind_identity_trace(
     Json(req): Json<BindIdentityTraceRequest>,
 ) -> Result<Json<ApiResponse<serde_json::Value>>, (StatusCode, Json<serde_json::Value>)> {
     let fd_table = crate::core::db::schema::table_name("face_detections");
+    let video_table = crate::core::db::schema::table_name("videos");
+    let video_table = crate::core::db::schema::table_name("videos");
     let id_table = crate::core::db::schema::table_name("identities");
     let history_table = crate::core::db::schema::table_name("identity_history");
 
@@ -854,6 +856,7 @@ pub async fn get_identity_traces(
 ) -> Result<Json<IdentityTracesResponse>, (StatusCode, String)> {
     let id_table = crate::core::db::schema::table_name("identities");
     let fd_table = crate::core::db::schema::table_name("face_detections");
+    let video_table = crate::core::db::schema::table_name("videos");
 
     let page = params.page.unwrap_or(1);
     let page_size = params.page_size.unwrap_or(20);
@@ -879,12 +882,13 @@ pub async fn get_identity_traces(
                       COUNT(*)::bigint AS frame_count,
                       MIN(fd.frame_number)::int AS first_frame,
                       MAX(fd.frame_number)::int AS last_frame,
-                      ROUND(MIN(fd.frame_number)::numeric / 25.0, 1)::float8 AS first_sec,
-                      ROUND(MAX(fd.frame_number)::numeric / 25.0, 1)::float8 AS last_sec,
+                      ROUND(MIN(fd.frame_number)::numeric / NULLIF(v.fps, 0)::numeric, 1)::float8 AS first_sec,
+                      ROUND(MAX(fd.frame_number)::numeric / NULLIF(v.fps, 0)::numeric, 1)::float8 AS last_sec,
                       ROUND(AVG(fd.confidence)::numeric, 4)::float8 AS avg_confidence
                FROM {} fd
+               LEFT JOIN dev.videos v ON fd.file_uuid = v.file_uuid
                WHERE fd.identity_id = $1
-               GROUP BY fd.file_uuid, fd.trace_id
+               GROUP BY fd.file_uuid, fd.trace_id, v.fps
                ORDER BY fd.file_uuid, fd.trace_id
                LIMIT $2 OFFSET $3"#,
         fd_table