Add clickhouse_keeper and default replica params

[jbuckmccready]

Description of the Change

Added Clickhouse Keeper and default replica parameters to config generation.

#64

[gallejesus]

I believe it should also include the Keeper nodes config (host & port) in config.xml.j2 so all CH nodes know where Keeper is, similar to what's doing with ZK

[jbuckmccready]

The Clickhouse Keeper config section only goes in the config for the nodes which are running keeper (not all the nodes like the zoo keeper config).

E.g., if you have a cluster of 4 data nodes and separately run 3 nodes with keeper on them then the zoo keeper config goes in all 7 nodes configs, but the keeper config only goes in the 3 keeper nodes.

The cluster nodes use the zoo keeper config to contact the keeper nodes (clickhouse keeper requires the zoo keeper config to be present).

[gallejesus]

Yes, I see there is an alternative config using <keeper> tags , so both configs are valid https://github.com/ClickHouse/ClickHouse/blob/master/tests/integration/test_alternative_keeper_config/test.py