Properly wait for postgres to finish spinning up instead of crashing out

[JuniPuff]

Helm chart version: 0.1.10
Store: Postgres

When using the helm chart and deploying a new instance of corndogs, when the DB is not up and ready to accept connections the main server pod crashes. After the DB is up, the server starts and runs the goose migrations just fine.