MBTI-PDF-RESULT-SNAPSHOT-SHELL: remove runtime pollutants from MBTI print snapshot shell

[fermatmind]

What changed

• Added a v3 print snapshot shell request contract in proxy.ts for /result/{id}/print?surface=mbti.result_page_snapshot.v3.
• Updated the localized layout to use the internal snapshot shell header and skip SiteChrome, header, footer, and CookieBanner while preserving providers/locale/font/runtime wrappers.
• Updated MbtiDesktopCloneShell snapshot mode to remove the sticky rail, ON THIS PAGE/TOOLS, final offers, checkout/unlock flows, share/retest/history/PDF/order/workspace actions, supplementary CTAs, recommended reads, and trailing footer-like nodes.
• Added/updated contract tests for proxy/layout scoping, Gotenberg print route markers, and MBTI desktop clone snapshot pollutant removal.
• Registered the PR-C train item and local validation state.

Why

PR-A established the v3 route/surface fail-closed contract and PR-B injected real desktop clone content. This PR-C removes ordinary web runtime pollutants from the v3 /print snapshot shell so content-ready pages are not blocked by header/cookie/footer/sidebar/tools/commerce UI.

Explicit contract answers

1. New shell: mbti.result_page_snapshot.v3 /print uses an internal header-driven PDF snapshot shell.
2. SiteChrome/Header/Footer/Cookie: not mounted in the snapshot shell.
3. Sidebar/ON THIS PAGE/TOOLS: not rendered in MBTI snapshotMode.
4. PDF/order/workspace/share/retest/unlock/final offer: not rendered in MBTI snapshotMode.
5. Ordinary /result/{id} pages: unchanged; the snapshot shell header is only set for v3 /print.
6. Backend/API: unchanged.
7. Final PDF visual quality: not guaranteed by this PR.

Validation

• pnpm exec vitest run tests/contracts/result-private-print-chrome.contract.test.ts tests/contracts/result-gotenberg-print-route.contract.test.ts tests/contracts/result-client-view-state.contract.test.tsx tests/contracts/mbti-desktop-shell-cta.contract.test.tsx -> pass, 4 files / 54 tests
• pnpm test:contract -> pass, 603 files / 3559 tests
• pnpm typecheck -> pass
• NEXT_PUBLIC_API_URL=https://api.fermatmind.com pnpm build -> pass
• ruby -e "require 'yaml'; YAML.load_file('docs/codex/pr-train.yaml'); puts 'yaml ok'" -> pass
• python3 -m json.tool docs/codex/pr-train-state.json >/dev/null -> pass
• git diff --check -> pass

Intentionally deferred

• No fap-api changes or backend PDF engine changes.
• No content import or CMS/storage mutation.
• No rendered PDF smoke automation; that is PR-D.
• No A4 pagination or visual polish; that is PR-E.
• No deploy, env change, cache clear, service reload, or production operation.

Repository rule impact

This is frontend product-code rendering/shell behavior for the private MBTI result print snapshot. It does not change CMS/content authority, SEO enumeration, public editorial content, sitemap authority, or backend-owned content surfaces.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: dcc689f89e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Record the reviewed head SHA in the train state

For this PR-train entry the state is marked committed, but the recorded commit_sha still points at 60914c25148a6f41088466dfcc904c827d6f05c4 while the commit under review is 17e7cedebf80d6c04397fa4b01136cffc21eb251. Any train reconciliation or dependency check that trusts this field will verify/report the wrong revision after this commit is pushed or squashed, so the ledger should be updated to the actual head SHA.

Useful? React with 👍 / 👎.