look to mapPartitions for performance increase

[ghukill]

After converting XSLT transformations that use pyjxslt to use rdd.mapPartitions (commit), and a relatively substantial performance increase, look to other areas where this could be applied:

• validation udfs in record_validation.py
• python transformation
  • move the temp ModuleType to the partition mapping
• Field mapping for ES, for index_mapper_handle() that comes from GenericMapper

Anything that is not row specific, can be removed from the UDF and applied at the partition level.

[ghukill]

Aside from instantiating index mapper only once per partition, as opposed to each record -- which has been implemented -- not seeing much potential for improvement from mapPartitions in field mapping.

[ghukill]

Applied where applicable, some definite performance gains (2-8x, depending).