fix(datafusion): keep hot runs in one session

[kumarUjjawal]

Summary

This fixes the partitioned DataFusion runner so hot runs are really hot. See apache/datafusion#21696

Before this change, the script started a new datafusion-cli process for every try. That made each try use a cold process, so tries 2 and 3 did not keep the warmed metadata cache.

What changed

• drop OS cache once before each query
• build one SQL file with create.sql and the same query 3 times
• run that file in one datafusion-cli session
• skip the setup Elapsed lines from create.sql
• keep failed later tries as null instead of mixing in setup timings

This follows the same main idea used by the DuckDB partitioned runner: one process per query, with repeated runs inside that process.

Thank you for review 🙏

Thank You for Your Contribution!

We appreciate your effort and contribution to the project. To ensure that your Pull Request (PR) adheres to our guidelines, please ensure to review the rules mentioned in our contribution guidelines:

ClickHouse/ClickBench Contribution Rules

Thank you for your attention to these details and for helping us maintain the quality and integrity of the project.

[CLAassistant]

All committers have signed the CLA.

[kumarUjjawal]

@qoega If I could get your eyes on this?