fix: window in scalar subquery returns wrong results

[xudong963]

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

Summary

Because we introduced single to inner join optimization, so we don't need to maintain the method rewrite_uncorrelated_scalar_subquery instead of rewriting to single join directly.

Then we can solve the issue15977.

Note: select count(distinct a) over (partition by 1) from t1; PARTITION BY 1 is a special syntax that actually means no partitioning, i.e., all rows are treated as a partition. So partition by (select b from t2) is not semantic if it returns more than one row but if returns only one row, it's valid, meaning no partitioning.

• fixes: Bug: subquery in partition by clause return wrong result #15977

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