Improve do_not_merge_across_partitions_select_final optimization.

[Avogar]

I hereby agree to the terms of the CLA available at: https://yandex.ru/legal/cla/?lang=en

Changelog category (leave one):

• Performance Improvement

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):
Add parallel select final for one part with level>0 when do_not_merge_across_partitions_select_final setting is 1.

Relates: #15938 and #11722

[azat]

AST fuzzer (UBSan) — ../src/Functions/FunctionDateOrDateTimeAddInterval.h:92:26: runtime error: signed integer overflow: -9223372036854775807 * 60 cannot be repr

#19466

[Avogar]

If I understand correctly, then length(thread_ids) > 1 is not correct, because there will be threads at least for:

• TCPHandler
• PipelineExecutor
• parts loading -

  ClickHouse/src/Storages/MergeTree/MergeTreeDataSelectExecutor.cpp

  Line 668 in f996b84

  /// Parallel loading of data parts.

Yes, you are right, I didn't think about it, thanks.

[Avogar]

@alexey-milovidov All checks passed, can I merge it?

[alexey-milovidov]

Yes, sure.