[Refactor] Manage Spark job parameter for different query

[dai-chen]

Is your feature request related to a problem?

Different query requires different Spark job submit parameters.

What solution would you like?

1. CREATE statement:
   a. Set execution-timeout-minutes to 0 so EMR-S won't time out the streaming job at background
   b. Disable DRA (Spark dynamic resource allocator) because it only works well for long-running batch job
   c. [TBD] Flint config that specifies which Env variable to populate to Flint metadata, such as EMR-S job info
2. SELECT statement

Also think about if any flexible way to manage these parameters in case of changes in future.

[penghuo]

• Replace Jar with Maven repository. https://github.com/opensearch-project/sql/blob/main/spark/src/main/java/org/opensearch/sql/spark/data/constants/SparkConstants.java. Depend on Implement FlintJob Logic for EMR-S opensearch-spark#52. Resolved by Add customized result index in data source etc #2212