Add checkpoint mandatory configuration

[dai-chen]

Description

Add new Flint config for checkpoint mandatory option. It is enabled by default which means all incremental refresh (CREATE statement with auto_refresh=true) must provide checkpoint_location. Doc: https://github.com/dai-chen/opensearch-spark/blob/add-mandatory-checkpoint-option/docs/index.md#configurations

TODO

Currently all Spark streaming job related validation happens when job start. For example, OS index is created even though job start failed due to missing checkpoint location. For this checkpoint validation, it can be solved by building streaming job early. However, other check on table/options maybe performed only when job start. Need to figure out how to validate early in general. Issue: #65

Example

spark-sql> CREATE SKIPPING INDEX ON ds_tables.http_logs
         > (status VALUE_SET)
         > WITH (
         >   auto_refresh=true
         > );
java.lang.IllegalStateException: Checkpoint location is mandatory for incremental refresh 
if spark.flint.index.checkpoint.mandatory enabled

# Currently create OS index succeeds. Will improve as TODO above.
spark-sql> DROP SKIPPING INDEX ON ds_tables.http_logs;

spark-sql> SET spark.flint.index.checkpoint.mandatory=false;
spark.flint.index.checkpoint.mandatory	false

spark-sql> CREATE SKIPPING INDEX ON ds_tables.http_logs
         > (status VALUE_SET)
         > WITH (
         >   auto_refresh=true
         > );

spark-sql> DESC SKIPPING INDEX ON ds_tables.http_logs;
status	int	VALUE_SET

Issues Resolved

#87

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.