Migrating to temporal workflows

[rustyrazorblade]

There's been some interest recently in using easy-cass-lab for production. At the moment it's not a great idea - there's a lot that's missing, and I don't really think it'll make sense to keep adding logic up for longer running processes. To do it right we need good workflow support. I think the path forward here would be to use Temporal.

Now, this is a little tricky, because I don't want to burden the regular user (me) with needing to set up temporal in AWS before getting started with ECL. However, for the user that would want to use ECL for a production cluster, that's exactly what we'd need.

I think the path forward is to be able to support two options.

1. Running Temporal in a local docker container (default)

The container would start automatically on up and be interacted with the same way Packer and Terraform are, although I'd expose local port(s) from the container, because Temporal would be a long running process.

2. Using an existing temporal environment (opt-in)

By leveraging an existing environment, we can deploy production-grade workflows that do all the common operations that are needed to run a Cassandra cluster.