Recently, AWS added session reuse to the Redshift Data API. It allows to, say, create a temporary table in one statement, and select from it in a subsequent one.
I think it would be useful that RedshiftDataOperator supports this new feature.

Decisions

• the hook and operator's database arguments now optional because when the session ID is provided, boto3 doesn't allow database to be specified. Because this becomes a named argument, I had to move it after sql. If users of the operator and hook rely on the position to pass the argument that could break things ⚠️ . Note that boto3 only requires Sql.
• the hook now returns an object containing the statement ID and session ID instead of just a statement ID. I choose a dataclass instead of a dict to make it harder to make a mistake when indexing the keys. If users use the hook in their tasks or custom operators, this could also break things ⚠️
• fixed the typo in the parse_statement_resposne method. It's a public one. So, again, there is a risk of breaking change. ⚠️
• boto3's validation error messages are not very useful when the session ID is a non-UUID string, or if no database, workgroup or session ID was provided, so I added checks in the hook.

To do

• update docs
• update changelog
• update and run examples
  • example_redshift.py
  • example_redshift_s3_transfers.py

^ Add meaningful description above
Read the Pull Request Guidelines for more information.
In case of fundamental code changes, an Airflow Improvement Proposal (AIP) is needed.
In case of a new dependency, check compliance with the ASF 3rd Party License Policy.
In case of backwards incompatible changes please leave a note in a newsfragment file, named {pr_number}.significant.rst or {issue_number}.significant.rst, in newsfragments.