Add more tests for Spark and BigQuery Also add script for running the tests on Spark. Also fix snake_case_to_proper_case in Spark. Also remove STRUCT columns from flatten_column output to match BigQuery. PiperOrigin-RevId: 596599316 Change-Id: Ib9043cd094019ac34a84112e5299d7dca6a37146