How to develop and test new features without affecting an already running bot? (aka. staging env)

[reliveyy]

We need to make a copy of following 4 points to have a totally seperated bot staging environment which could be used to test some new features. And it's open to discuss what are the minimum requirements which need to run another bot testing instance. Or how do we run bot locally when developing new features?

1. Only one #xud-docker-bot channel
2. Only one Travis-CI project (ExchangeUnion/xud-docker)
3. Only one GitHub webhook (xud-docker, xud, market-marker-tools, boltz-lnd)
4. Only one DockerHub group (ExchangeUnion)