refactor(query): experimental aggregate final with recursive spill support

[dqhl76]

I hereby agree to the terms of the CLA available at: https://docs.databend.com/dev/policies/cla/

Summary

refactor(query): add experiment aggregate final to support recursive spill

introduce enable_experiment_aggregate_final setting, default is false

• Temporarily enabled (true) for CI testing convenience; will be turned off before merge

This PR aims to address the OOM issue occurring during the final aggregate stage.

Key changes

• Introduces recursive repartitioning of large partition payloads until they are sufficiently small.
• Use stream partition pattern to control the spill and restore payload size

Planned follow-ups

• Update benchmarks to enable this setting by default.
• Introduce a setting option to control the maximum recursive spill depth.

Tests

• Unit Test
• Logic Test
• Benchmark Test (TODO)
• No Test - Explain why

Type of change

• Bug Fix (non-breaking change which fixes an issue)
• New Feature (non-breaking change which adds functionality)
• Breaking Change (fix or feature that could cause existing functionality not to work as expected)
• Documentation Update
• Refactoring
• Performance Improvement
• Other (please describe):

---

This change is 

[github-actions]

Docker Image for PR

• tag: pr-18866-3f17e07-1761463276

note: this image tag is only available for internal use.

[github-actions]

ClickBench Report

• hits: https://benchmark.databend.com/clickbench/pr/18866/18814095310/hits.html
• internal: https://benchmark.databend.com/clickbench/pr/18866/18814095310/internal.html
• load: https://benchmark.databend.com/clickbench/pr/18866/18814095310/load.html
• tpch: https://benchmark.databend.com/clickbench/pr/18866/18814095310/tpch.html