[Eeboade]

Summary

The web frontend has no camera-permission gate or QR scanner: QRCodeDisplay only renders a code, and the send flow has no scanner. I added a reusable CameraPermissionDialog (ignition-pay-frontend/components/camera-permission-dialog.tsx) built on the shared Base UI Dialog, so it gets role=dialog, aria-modal, a focus trap and Escape to close. It has a labelled title and description, and a role=status aria-live=polite region that announces the permission state (prompt, requesting, granted, denied, unavailable). The primary 'Allow camera' / 'Try again' button is auto-focused and works from the keyboard, and Cancel closes the dialog. I did not wire it into QRCodeDisplay or SendPage, because there is no scanner there yet. A future scanner should render this dialog and drive its state prop from getUserMedia results. I couldn't run anything, so none of this has been run.

Changed files

• ignition-pay-frontend/components/camera-permission-dialog.tsx
• ignition-pay-frontend/tests/camera-permission-dialog.test.tsx

Test plan

New test ignition-pay-frontend/tests/camera-permission-dialog.test.tsx checks the dialog's name, description and aria-modal, the live status text, the allow button click, and that the allow button is hidden when no camera exists. CI should run the frontend vitest suite, plus lint and typecheck.

This PR was created as a draft by the issue solver bot. It will remain a
draft until repository CI passes.

Closes #859

[drips-wave]

@Eeboade Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits