Support janusgraph.sh options to use HBase and/or Solr as backend

The current janusgraph.sh starts a Cassandra and ES together with the gremlin server for convenient use.
Let's add options to be able to fork and use HBase and/or Solr instance as well.

Start by reading janusgraph.sh and tracing how it currently starts Cassandra, Elasticsearch, and the Gremlin server. Identify the existing startup option handling and determine how HBase and Solr instances should be selected. Done means the script can independently enable HBase and/or Solr backends while retaining the current startup behavior.

Written by the indexing model from the issue text.