test: add React Native scene notification E2E (MBL-2303)

[Shahroz16]

Summary

• generate a React Native 0.88 UIScene host against the exact wrapper under test
• tap real simulator notifications from terminated and warm app states with Maestro
• assert each exact destination reaches React Native Linking
• run nightly/manual on the Xcode 27 runner, plus pull-request validation when the E2E harness changes

The runner waits for Maestro to reach the Home screen before injecting each notification, so slower hosted startup cannot race the banner. No app-side test hook or workspace credential is added.

Validation

• local Xcode 26.6 / iOS 26.5 clean host build passed
• terminated notification tap passed
• warm notification tap passed
• ESLint, TypeScript, Jest, shell/Ruby/YAML/JSON syntax, and actionlint passed; actionlint only reports the repository custom xcode-27 runner label

This validates simulator presentation, tap handling, native buffering, and client routing. It does not claim real APNs/FCM token registration or Customer.io backend sent/delivered/opened evidence.

[github-actions]

Sample app builds 📱

Below you will find the list of the latest versions of the sample apps. It's recommended to always download the latest builds of the sample apps to accurately test the pull request.

---

• Android APN: 643.15.0 (29796293)
• iOS APN: 643.15.0 (29796293)

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, have a team admin enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 98783b2. Configure here.}

[mrehan27]

Approving. The design is sound and the cold flow is genuinely discriminating — terminating the app before the tap means a terminated-tap really does exercise the buffer-until-JS-ready path, so this guards the feature rather than just reaching the destination.

The comments below are all the same theme: the hardening that landed in customerio/customerio-flutter#400 hasn't been carried over to this lane. None of it blocks — the lane is green and the coverage is real — but the pinning one is worth doing before merge.