improved deployment process

[josefaidt]

Issue #, if available:

resolves #509
blocked by #495
blocked by #514
blocked by #513
blocked by #512

Description of changes:

This PR aims to revise the deployment process of the Discord Bot project. See #509 for details. In short, every PR merged to main will deploy to production. The new process will run tests on the live environments post-deployment to ensure the app functions as expected, and will stop deployment to production if e2e test fail in pre-prod (next)

• creates reusable "configure-node" action that configures Node.js with corepack enabled and custom pnpm cache
• creates "failed-release-notification" workflow to create an issue when a deployment fails. Sets the foundation to extend to other platforms like Slack.
• removes "create-release" workflow in favor of bumping versions based on PR "semver" labels
• modifies "release" action to run on every PR merge (rather than looking at PR branch name and author)
• modifies "release-env" to run e2e tests on live environment (e.g. verify guild cookies are handled as expected, auth flow verification TBD)

Things that could be improved:

• release failure notification currently only creates an issue. it should also message via Slack, etc.
• release failure notification does not have context of deployment. it would be nice to provide release context such as new version, job/step at which it failed, etc. when creating the issue
• automate viewing of Playwright test results?
• ensure PRs only have 1 database migration file to ensure reliability of migrations during deployment

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.