dgraph-io/dgraph ? reverse-engineered prompt

Build me a production ready distributed graph database in Go that can run as a cluster and handle real time queries at scale.

I want it to support GraphQL style querying, JSON and protocol buffers over HTTP and gRPC, and give me ACID transactions, consistent replication, and fast reads. It should work well for graph heavy data, with traversals, filters, aggregations, and full text search, and it should feel simple enough to use like a modern GraphQL database. Please make it suitable for Linux amd64 and arm64, and make Docker the easiest way to run it locally. A quick standalone setup would be great too.

If you need to look up current docs online, do that. Keep the code clean, production minded, and include the basic commands or examples needed to build and start it from source.