dev environments never die

[schdief]

leads to overload and downtime in production, so:

• build only for PRs and master branch
• ketchup UAT stage as yaml release pipeline
• think about stage naming for UAT
• add stage PROD to ketchup release pipeline
• release pipeline as .yaml for gui
• seems like implicit downloading fails us at cypress (so deactivate it?), but also find out, why explicit download doesn't work: Can't run because no spec files were found.
• check build triggers, seems like ketchup runs, although only gui was changed!?
• disable old pipelines in az devops
• PR environments are all named "merge" in helm, therefore they override each other, try source.Branch instead of source.BranchName (need to replace / with _ or just take out the PR#) -> https://docs.microsoft.com/en-us/azure/devops/pipelines/build/variables?view=azure-devops&tabs=yaml
• set up job to helm delete a PR release after it is closed (with Github Actions?)
• try out cache in gui pipe for npm install: https://docs.microsoft.com/en-us/azure/devops/pipelines/caching/?view=azure-devops
• check PROD deployment (requires temporary master branch filter removal)