[Analytics] Report additional bulk analytics job stats for instrumentation

Currently, the Cassandra bulk analytics library supports a "dialHome" API to publish some initial job metadata, which in its current form, is redirected to a log. The intention behind this is to allow custom implementations that can utilize these summarized stats for instrumentation or reporting of client behavior.

This task is meant to enhance this API to allow for additional job metadata to be published both at the Spark executor level and at the task levels to gather stats such as "success/failure", "number of rows written/read", "failure reason" etc.