document the deployment procedure please please please #262

We need tod ocument the deployment procudere because:

1. Every second deployment fails because of some human error
2. Only oren really knows how it works

So we need step-by-step documentation on:

• how to add a new dao to an existing subgraph version
• how to deploy a new subraph version

These steps involve:

• copy an exiting directory or git checkout (?)
• edit docker-compose and assign new ports
• star the dcker cotnainers
• editing the necessary files for subgraph deploymnet
• npm run deploy
• check if server is done indexing
• configure nginx and reload it

We can document this in the https://github.com/daostack/devops/ repo so it will not be public.

the procedure is quite complex, perhaps more than necessary.

My impression is that deployment (of a new rinkeby or mainnet subgraph version) has become very cmplex, with file changes (apparently) in these files that influence the results:

• the .env file
• the daos/ directory
• the ops/mappings.json
• the ops/mapping.json container
• the migration.json from daostack/migration
• the docker-compose file

If that is really the case, perhaps it is also a good idea to reorganize this