This module starts both the gRPC server and the HTTP server backed by a single MLeap executor, so you can serve models over both protocols from one process.
The gRPC server is provided by mleap-grpc-server and the HTTP server is
provided by mleap-spring-boot. Both share the same MleapExecutor, so a model
loaded through one interface is available through the other.
The entry point is ml.combust.mleap.serving.RunServer. See that file for the
available configuration and startup details.