test(test-app): add a perf lab with seeded React render problems

examples/test-app/README.md

@@ -18,6 +18,7 @@ It is intentionally small, but each surface is dense with durable accessibility
 - `Settings`: switch rows, accordion content, loading and error states, retry flow, destructive-confirm modal
 - `Automation lab`: long-press, alert-result, app-event, app-state, appearance, orientation, permission-recovery, log canaries, a flattened (`accessible={true}`) text input, and an Apple Pay sheet hosted in `com.apple.PassbookUIService`
 - `WebView accessibility`: a deterministic semantic fixture plus live websites with varied HTML for native accessibility snapshot verification
+- `Perf lab`: seeded JavaScript render problems for `react-devtools profile`, one per scenario (see [Perf lab](#perf-lab))
 
 Navigation uses Expo Router native bottom tabs, so the tab bar itself is also part of the test surface.
 
@@ -234,6 +235,32 @@ installed development build is connected to that server. For Android local
 device/emulator runs, also run `adb reverse tcp:8082 tcp:8082` when the device
 needs host port forwarding.
 
+## Perf lab
+
+`agent-device-test-app://perf-lab?scenario=<id>` opens one seeded render problem. **Run** (`perf-run`) replays the slow interaction as a fixed burst and sets `perf-status` to `done`. Profile it with:
+
+```bash
+agent-device open "agent-device-test-app://perf-lab?scenario=memo-defeated" --platform android
+agent-device react-devtools profile start
+agent-device click 'id="perf-run"'
+agent-device wait 'id="perf-status" text="done"' 10000
+agent-device react-devtools profile stop
+```
+
+`profile stop` should report this finding first:
+
+| Scenario | Seeded problem | Expected finding |
+| --- | --- | --- |
+| `expensive-render` | `PriceHistoryChart` recomputes its trend on every render | `slow-render PriceHistoryChart` |
+| `memo-defeated` | `ContactList` passes a new callback and style object to memoized rows | `memo-defeated ContactRow` |
+| `context-cascade` | A fast-changing `unread` counter shares one context with the stable user name that badges read | `context-cascade AvatarBadge` |
+| `parent-cascade` | Typing re-renders every unmemoized result row | `parent-cascade ResultRow` |
+| `render-storm` | `LiveClock` sets state every 16 ms | `render-storm LiveClock` |
+| `effect-chain` | `FilteredContacts` copies derived state in an effect | `effect-chain FilteredContacts` |
+| `eager-list` | 1,200 tiles mount in one commit instead of a virtualized list; every Run mounts them afresh | `eager-mount ProductTile` |
+
+`src/screens/PerfLabScreen.tsx` imports only React and React Native, so it can be copied into a bare React Native app to check the profiler on other React Native versions.
+
 ## Local Agent Device suites
 
 The repo includes two local suites for iterating on the fixture app:

examples/test-app/app/_layout.tsx

@@ -29,6 +29,7 @@ function RootLayoutContent() {
         <Stack.Screen name="snapshot-depth" />
         <Stack.Screen name="deep-tree" />
         <Stack.Screen name="ticker" />
+        <Stack.Screen name="perf-lab" />
         <Stack.Screen name="product/[productId]" />
         <Stack.Screen name="webview" />
       </Stack>

examples/test-app/app/perf-lab.tsx

@@ -0,0 +1,9 @@
+import { useLocalSearchParams } from 'expo-router';
+
+import { isPerfScenario, PerfLabScreen } from '../src/screens/PerfLabScreen';
+
+export default function PerfLabRoute() {
+  const { scenario } = useLocalSearchParams<{ scenario?: string }>();
+  const initialScenario = isPerfScenario(scenario) ? scenario : undefined;
+  return <PerfLabScreen initialScenario={initialScenario} key={initialScenario} />;
+}