docs: add standing GUI-testing guidelines and scenario library

[Claudius-Maginificent]

Why this PR exists

• Problem: Verifying a change against the real, compiled app on a real display is a distinct testing tier from kittest (in-process harness, no display) and backend-e2e (calls BackendTasks directly, no UI) — but there's no standing home for the guidance and safety rules that kind of testing needs, so every agent/session re-derives them from scratch (launch recipe, credential handling, fund-movement safety, isolated-data-dir discipline).
• What breaks without it: knowledge from one live-GUI test run (e.g. exact navigation steps, timing gotchas, safety constraints like "use the owner-key withdrawal path since it forces a consensus-verified destination") evaporates at the end of the session instead of compounding into a reusable, ever-more-reliable scenario library.
• Blocking relationship: none — pure documentation addition, stacked on docs/platform-wallet-migration-design (feat: platform-wallet backend rewrite (spec + implementation) #860) since that's the current active integration branch.

What was done

• New docs/gui-testing/README.md: when to reach for this tier vs. kittest/backend-e2e, non-negotiable safety rules (isolated data dir, never touch an already-running instance, credential handling by env-var name only — never hardcode secrets, testnet-by-default, capped fund movement, prefer consensus-forced destinations, always check logs for a panic even when the UI looked fine), and the scenario file format/index.
• New docs/gui-testing/scenarios/TEMPLATE.md: the template for individual scenario files, written at the level of "what to look for" rather than pixel coordinates so scenarios survive minor UI changes.
• CLAUDE.md: two new pointers — one under Testing (agents about to run a live GUI test find it), one under Documentation (consistent with how docs/ai-design, docs/personas, etc. are catalogued). Deliberately not date-grouped under docs/ai-design — this is standing, reusable practice meant to be discovered on every future GUI-verification task, not a point-in-time design record.

Testing

Documentation-only change — no code paths affected. cargo fmt/clippy/tests unaffected (verified no source files touched).

Breaking changes

None.

Checklist

• N/A — docs-only, no tests/clippy affected
• Reviewed by a human

Attribution

_{🤖 Co-authored by Claudius the Magnificent AI Agent}

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

🗂️ Base branches to auto review (2)

• master
• v1.0-dev

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: e87c5494-b729-433f-a9bf-fc29948b58c1

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch docs/gui-testing-guidelines

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}