[CQLReplicator on Glue] Update CQLReplicator.scala for memorydb, parquet, and opensearch

[nwheeler81]

Proposed changes:

• Replace sparkSession.read.parquet in dataReplicationProcess() and keysDiscoveryProcess() to glueContext.getSourceWithFormat with DISK_ONLY
• Add boolean data type in rowToStatement
• Add replicationPointInTime to replicate data from a specific data point
• Add ClientConfiguration with retries for the S3 client
• Add aggregated stats for dataReplicationProcess(), read count.json, aggregate, update count.json. Proposed new structure: { "tile": 0, "primaryKeys": value "updatedPrimaryKeys": value "insertedPrimaryKeys": value "deletedPrimaryKeys": value "updatedTimestamp": value }
• Updated cqlreplicator to return new stats --state stats
• Update README.MD

[nwheeler81]

Left changes only for MemoryDB and OpenSearch connectors:

• Replace sparkSession.read.parquet in dataReplicationProcess() and keysDiscoveryProcess() to glueContext.getSourceWithFormat with DISK_ONLY
• Add replicationPointInTime to replicate data from a specific data point
• Add ClientConfiguration with retries for the S3 client
• Add aggregated stats for dataReplicationProcess(), read count.json, aggregate, update count.json. Proposed new structure: { "tile": 0, "primaryKeys": value "updatedPrimaryKeys": value "insertedPrimaryKeys": value "deletedPrimaryKeys": value "updatedTimestamp": value }
• Updated cqlreplicator to return new stats --state stats