Use Trigger.dev v4 API keys and access tokens correctly across dev, staging, preview and production

Documented steps

1. Read the key model first: https://trigger.dev/docs/apikeys.
2. For triggering and Management API calls from backend code set `TRIGGER_SECRET_KEY` (auto-detected), or configure it explicitly: `import { configure } from "@trigger.dev/sdk"; configure({ secretKey: process.env.TRIGGER_SECRET_KEY });`
3. Match key prefix to environment: `tr_dev_` (development), `tr_stg_` (staging), `tr_prod_` (production).
4. Confirm the runtime environment inside a task via `ctx.environment.slug` and `ctx.environment.type` (`DEVELOPMENT` | `STAGING` | `PREVIEW` | `PRODUCTION`) (https://trigger.dev/docs/deploy-environment-variables).
5. Never hardcode secret keys — load them from environment variables or a secret manager.
6. For CI/CD deploys authenticate with `TRIGGER_ACCESS_TOKEN` rather than the interactive `npx trigger.dev@latest login` (https://trigger.dev/docs/cli-deploy-commands).
7. For self-hosted instances also set `TRIGGER_API_URL`, or the SDK/CLI talks to Trigger.dev Cloud instead.
8. For any browser code mint a Public Access Token or Trigger Token server-side and pass only that to the client (https://trigger.dev/docs/realtime/auth).
9. For isolated local work use dev branches via the CLI `--branch` flag (https://trigger.dev/docs/deployment/dev-branches).