AI: Quite llama-server on app force quit

Author: vdavid

apps/desktop/src-tauri/src/ai/CLAUDE.md

@@ -80,6 +80,9 @@ Frontend                    manager.rs              process.rs / download.rs / c
 **Decision**: `SIGTERM` then 5s wait then `SIGKILL` for process shutdown.
 **Why**: llama-server may be mid-inference holding GPU memory. `SIGTERM` gives it a chance to release resources cleanly. The 5s timeout prevents hanging on app quit if the server is stuck.
 
+**Decision**: `shutdown()` called from both `on_window_event` (CloseRequested/Destroyed) and `RunEvent::Exit`.
+**Why**: `on_window_event` handles normal quit, but force-quit/crash/SIGTERM bypass it. `RunEvent::Exit` fires on app-level exit regardless of how it was triggered. `shutdown()` is idempotent (`child_pid.take()` returns `None` on subsequent calls), so double-calling is safe.
+
 **Decision**: Context window (`-c 4096`) explicitly set on llama-server.
 **Why**: Without `-c`, llama-server defaults to the model's trained max context (256K for Ministral), creating a ~27 GB KV cache. Folder suggestions need at most 2K context. 4K is generous and keeps memory under ~400 MB.
 

apps/desktop/src-tauri/src/lib.rs

@@ -872,6 +872,13 @@ pub fn run() {
                 network::mdns_discovery::stop_discovery();
             }
         })
-        .run(tauri::generate_context!())
-        .expect("error while running tauri application");
+        .build(tauri::generate_context!())
+        .expect("error while building tauri application")
+        .run(|_app, event| {
+            if let tauri::RunEvent::Exit = event {
+                ai::manager::shutdown();
+                #[cfg(any(target_os = "macos", target_os = "linux"))]
+                network::mdns_discovery::stop_discovery();
+            }
+        });
 }