[GLUTEN-7800][VL] Add config for max reclaim wait time to avoid dead lock when memory arbitration

[Yohahaha]

What changes were proposed in this pull request?

Add config spark.gluten.sql.columnar.backend.velox.reclaimMaxWaitMs to avoid dead lock when memory arbitration has bug, default timeout is 60min.

[github-actions]

Thanks for opening a pull request!

Could you open an issue for this pull request on Github Issues?

https://github.com/apache/incubator-gluten/issues

Then could you also rename commit message and pull request title in the following format?

[GLUTEN-${ISSUES_ID}][COMPONENT]feat/fix: ${detailed message}

See also:

• Other pull requests

[github-actions]

Run Gluten Clickhouse CI

[github-actions]

#7800

[Yohahaha]

@zhztheplayer would you help take a look?

[zhztheplayer]

Why we need to include SharedArbitrator.h? Thought we should avoid this kind of dependency. One of the reason is, code of the shared arbitrator is usually updated frequently so we may easily find our code doesn't work after a Velox rebase.

[Yohahaha]

I just want reuse velox::memory::SharedArbitrator::ExtraConfig, would you give some input?

[zhztheplayer]

Only one question about the header dependency. Thanks.

[zhztheplayer]

@Yohahaha

A bit dramatical that upstream Velox really updated the function name in SharedArbitrator.h...

See facebookincubator/velox@62b0a12#diff-00e1979427f8bb6a5bda4f62c02e98dfd2cc0bdadb9219772e667946ffe1421dR111

It causes main CI build failure and let's revert this PR