fix: tpcds spill test fail

[forsaken628]

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

Summary

• hash_join_desc.is_null_equal is not handled correctly, which causes Q10 Q35 Q38 Q87 to fail when spilling is enabled, this issue is fixed.
• Aggregation spilling and exchange are incompatible, which causes Q23 Q41 to fail when spilling is enabled and in cluster mode, but this is not yet fixed. panicked at src/query/service/src/pipelines/processors/transforms/aggregator/new_transform_partition_bucket.rs:220 @zhang2014 is working on refactoring
• The io performance of the ci environment is very poor, it takes about 900s to run a full tpcds with spilling enabled and in standalone mode, so only a third of the tpcds are randomized at a time.

part of #17450

Tests

• Unit Test
• Logic Test
• Benchmark Test
• 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