These are a handful of data structures and behaviors that can be combined to provide flexible
processing of Kafka messages.

SaramaMessageConsumer is tightly focused on the consumption of Kafka messages. **This is the
interface that other packages will want to implement in order to process Kafka messages.** Being
a simple interface, it can easily be extended to provide additional behavior, such as
retries. One such example is implemented, the NTimesRetryingConsumer.

SaramaConsumerGroupHandler implements sarama.ConsumerGroupHandler. The handler is controlled by
the consumer group, and is responsible for managing the lifecycle of a SaramaMessageConsumer. It
can be extended if the need for more lifecycle support is required.

The SaramaEventsConsumer combines a SaramaMessageConsumer and a SaramaConsumerGroupHandler
providing them with a simple lifecycle API (the Run() method). This allows other systems (namely
platform) to use a very thin adapter layer to control the lifecycle of the consumer.

BACK-2554

Have the basics there, but need to:

+ write tests
+ test it
+ manage consumers for the other topics, whether here or somewhere else