HDDS-11721. Fix container export failure when container scanner is running on a schema v2 container

[ptlrs]

What changes were proposed in this pull request?

Container export operations for Schema v2 containers can fail when a scan is simultaneously in progress. This occurs because the scanner maintains an active database (Db) handle, preventing the export process from clearing the Db handle cache.

Previously, the export process attempted to clear this cache (a feature introduced in HDDS-3363). However, with the Db handle still held by the scanner, this cache-clearing instruction would fail, leading to the overall export failure.

To address this, the export process has been updated to no longer attempt to clear the Db handle cache, allowing for successful container exports even when a scan is ongoing.

A unit test has been added which reproduces the failure conditions.

What is the link to the Apache JIRA

https://issues.apache.org/jira/browse/HDDS-11721

How was this patch tested?

CI: https://github.com/ptlrs/ozone/actions/runs/16429140937

Flaky test 10x10 for:
TestKeyValueContainer
TestDecommissionAndMaintenance

[ptlrs]

Hi @errose28 @Tejaskriya @adoroszlai can you please review this PR?

[Tejaskriya]

Thanks for the patch and the extensive test cases added @ptlrs !
I was just thinking if it is possible to introduce another utility method in BlockUtils that would try to clear out the cache, and ignore in case of failure. This way in the majority of cases where export+scanner is not happening simultaneously, we will be clearing the cache. But if they are happenening simultaneously, the cache clearing will fail, but the export operation itself wouldn't as we won't pass the exception out of the utility method introduced.
What do you think?