[FEATURE] Add Flint config for env to persist in metadata

[dai-chen]

Is your feature request related to a problem?

In https://github.com/opensearch-project/opensearch-spark/pull/58/files, we hardcoded the runtime env variable name and persisted them to Flint metadata. The hardcoding makes Flint codebase aware of underlying runtime env like EMR-S and inflexible if we want to persist

What solution would you like?

Provide another Flint configuration spark.flint.index.metadata.env. Example: with --conf spark.flint.index.metadata.env=envKey1,envKey2,..., Flint extension will persist these env variables to metadata when create index.