Add support for VARBYTE type

[aajisaka]

Problem

Fails to read from Redshift table if the table has VARBYTE type column

An error occurred while calling o156.load. : java.sql.SQLException: Unsupported type -4
    at io.github.spark_redshift_community.spark.redshift.data.JDBCWrapper.getCatalystType(JDBCWrapper.scala:445)
    at io.github.spark_redshift_community.spark.redshift.data.JDBCWrapper.resolveTable(JDBCWrapper.scala:353)
    at io.github.spark_redshift_community.spark.redshift.RedshiftRelation.$anonfun$schema$1(RedshiftRelation.scala:74)
    at scala.Option.getOrElse(Option.scala:189)
    at io.github.spark_redshift_community.spark.redshift.RedshiftRelation.schema$lzycompute(RedshiftRelation.scala:68)
    at io.github.spark_redshift_community.spark.redshift.RedshiftRelation.schema(RedshiftRelation.scala:67)
    at org.apache.spark.sql.execution.datasources.DataSource.resolveRelation(DataSource.scala:460)
    at org.apache.spark.sql.execution.datasources.DataSource.resolveRelation(DataSource.scala:345)
    at org.apache.spark.sql.DataFrameReader.loadV1Source(DataFrameReader.scala:278)
    at org.apache.spark.sql.DataFrameReader.$anonfun$load$1(DataFrameReader.scala:210)
    at scala.Option.getOrElse(Option.scala:189)
    at org.apache.spark.sql.DataFrameReader.load(DataFrameReader.scala:210)

Testing

Built from source, add the jar in Glue 5.0, and ran a Spark job to write binary data to Redshift table and read them from the same table.

I didn't update the unit tests, but I'm willing to work on that. Please let me know how I can efficiently implement unit tests for the new type.

Notes

• Requires Redshift JDBC v2.1.0.1 or upper: https://github.com/aws/amazon-redshift-jdbc-driver/blob/master/CHANGELOG.md#v2101-2021-09-04
• I had to set tempformat to PARQUET instead of AVRO (default). Otherwise it fails with "The hex string size for Varbyte must be an even number" error. It seems Redshift LOAD from Avro doesn't support binary type.

[aajisaka]

@bsharifi Would you review?