Batching and localdc support

[patitapaban19]

Added support for the following:

1. scylladb.max.batch.size
2. scylladb.loadbalancing.localdc And documentation improvements for CONFIG.md.

[tarzanek]

see https://docs.datastax.com/en/drivers/java/3.7/com/datastax/driver/core/BatchStatement.html#add-com.datastax.driver.core.Statement-

• it describes all options that will be overridden by batch setting, so you will need to setup them on batch rather than on every statement in case batching is used (I think performance wise you might see that in some cases not using batching will be faster than in others)

[tarzanek]

for the timestamp issue I think we should open a talk to review the design from where I stand it looks like we should keep timestamp of every single record from kafka(as source of truth) so I guess the user should know when he starts using batching, that his timestamps will get the same or another solution is to add a specific kafka timestamp column