polish the rendering a bit, add a macro

examples/bouncing_ball.rs

@@ -32,6 +32,7 @@ impl Ball {
     }
 }
 
+#[instrument]
 fn main() {
     // Start the telemetry window
     teleprof::start();
@@ -49,7 +50,7 @@ fn main() {
     println!("- Flame graph rendering (all depth-N spans on row N)");
     println!("- instrument_calls macro for automatic call instrumentation");
     println!("- Runtime deduplication preventing double-wrapping");
-    println!();
+    println!(); 
 
     let mut window = Window::new(
         "Bouncing Ball",
@@ -102,6 +103,7 @@ fn main_frame(
             teleprof::set_thread_name(format!("ColorPicker-{}", id));
             pick_new_color(ball_clone);
         });
+        let _ = COLOR_PICKER_COUNTER.fetch_update(Ordering::Relaxed,Ordering::Relaxed,|cnt| Some(cnt -1));
     }
     
     // Render - also instrumented, so dedup will handle it
@@ -118,7 +120,7 @@ fn main_frame(
     }
 }
 
-#[instrument]
+
 fn update_physics(ball: &Arc<Mutex<Ball>>, dt: f32) -> bool {
     let mut ball = ball.lock().unwrap();
     
@@ -163,7 +165,6 @@ fn pick_new_color(ball: Arc<Mutex<Ball>>) {
 }
 
 // Using instrument_calls here too to see the breakdown of rendering
-#[instrument_calls]
 fn render(ball: &Arc<Mutex<Ball>>, framebuffer: &mut [u32]) {
     clear_background(framebuffer);
     draw_ball(ball, framebuffer);
@@ -203,7 +204,6 @@ fn submit_frame() {
     thread::sleep(Duration::from_millis(2));
 }
 
-#[instrument]
 fn print_status(ball: &Arc<Mutex<Ball>>, frame: u32) {
     let ball = ball.lock().unwrap();
     println!(