[FEATURE] Improving BI results retrieval with Cloud Fetch

### Code of Conduct

- [X] I agree to follow this project's [Code of Conduct](https://www.apache.org/foundation/policies/conduct)

### Search before asking

- [X] I have searched in the [issues](https://github.com/apache/incubator-kyuubi/issues?q=is%3Aissue) and found no similar issues.

### Describe the feature

https://databricks.com/blog/2021/09/08/new-performance-improvements-in-databricks-sql.html

### Motivation

Improving BI results retrieval with Cloud Fetch

### Describe the solution

>Once query results are computed, the last mile is to speed up how the system delivers results to the client – typically a BI tool like PowerBI or Tableau. Legacy cloud data warehouses often collect the results on a leader (aka driver) node, and stream it back to the client. This greatly slows down the experience in your BI tool if you are fetching anything more than a few megabytes of results.

> That’s why we’ve reimagined this approach with a new architecture called Cloud Fetch. For large results, Databricks SQL writes results in parallel across all of the compute nodes to cloud storage, and then sends the list of files using pre-signed URLs back to the client. The client then can download in parallel all the data from cloud storage. We are delighted to report up to 10x performance improvement in real-world customer scenarios! We are working with the most popular BI tools to enable this capability automatically.

by databricks

### Additional context


### Are you willing to submit PR?

- [ ] Yes I am willing to submit a PR!

Start with the linked Databricks SQL performance article and review how Kyuubi currently delivers query results to BI clients. Define how large results would be written to cloud storage, exposed through pre-signed URLs, and downloaded in parallel; the issue names no source files or tests, so completion criteria need to be established.

Written by the indexing model from the issue text.