feat: Provide a way to terminate the Kaskada process

[kerinin]

Summary Currently we suggest creating Kaskada instances for the CLI with:

./kaskada-manager 2>&1 > manager.log 2>&1 &
./kaskada-engine serve > engine.log 2>&1 &

The only way to stop these processes is with kill. Make it easier to stop these processes, possibly with a command like kaskada-cli terminate.

[epinzur]

idea add grpc endpoints that shutdown the services... but somehow don't allow access to those when running in production/kubernetes.