Wrap necessary method for changing env vars in run script

Our run script, https://github.com/dssg/matching-tool/blob/master/scripts/run does not have an easy way to change the environment variable in a container, or at least how Flask gets certain environment variables. The container needs to be recreated, not just stopped. But we don't need to rebuild. One way to do this is 'docker-compose down', but that brings down all containers. What if the matcher is running and we don't want to bug it? We can do 'docker-compose stop ' followed by 'docker-compose rm ', and then when it is recreated on 'docker-compose start ' it should have the fresh environment variables in Flask. We should couple the 'stop' and 'rm' in one command, or perhaps even combine 'stop', 'rm', and 'start' for a single container in one command. Maybe we call this command 'recreate'?