feat(query): support left plan's from clause contains subquery

[forsaken628]

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

Summary

SubqueryRewriter::flatten_plan rewrites a correlated subquery as a join. A rewrite involves a subquery plan and an outer plan.
The previous implementation didn't pass in the outer plan but got it indirectly from the metadata, so there were a lot of limitations, but in this PR, it passes in the outer plan directly.

Currently only DummyTableScan, ConstantTableScan, Scan, EvalScalar, Limit, Sort, Filter, Join, Aggregate are supported in outer plan.

#17476 (comment) cc @KKould

• fixes: bug: subquery with union cause panic #17476

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 

[github-actions]

Docker Image for PR

• tag: pr-17621-14eebfd-1742520259

note: this image tag is only available for internal use,
please check the internal doc for more details.

[github-actions]

ClickBench Report

• hits: https://benchmark.databend.com/clickbench/pr/17621/13982319773/hits.html
• internal: https://benchmark.databend.com/clickbench/pr/17621/13982319773/internal.html
• load: https://benchmark.databend.com/clickbench/pr/17621/13982319773/load.html
• tpch: https://benchmark.databend.com/clickbench/pr/17621/13982319773/tpch.html

[forsaken628]

bug reproduce on v1.2.707

CREATE OR REPLACE TABLE employees AS
SELECT * FROM (
    VALUES
        (1, 1, 50000),
        (2, 2, 60000),
        (3, 3, 70000)
) AS employees(employee_id, name, salary);

CREATE OR REPLACE TABLE departments AS
SELECT * FROM (
    VALUES
        (101, 1, 1),
        (102, 2, 2),
        (103, 3, 3)
) AS departments(department_id, department_name, manager_id);

CREATE OR REPLACE TABLE salaries AS
SELECT * FROM (
    VALUES
        (1, 50000),
        (2, 60000),
        (3, 70000),
        (4, 80000),
        (5, 90000)
) AS salaries(employee_id, salary);

SELECT 
    e.employee_id, 
    d.department_name
FROM 
    employees e
INNER JOIN 
    departments d 
ON 
    e.employee_id = d.manager_id
WHERE 
    e.salary > ALL (
        SELECT salary
        FROM salaries
        WHERE employee_id <> e.employee_id
    )
ORDER BY employee_id;

---

FIXED

[forsaken628]

bug reproduce on v1.2.707

create or replace table tab2 as select * from (values (1, 10), (1, 20), (2, 30)) tab(n, m);
create or replace table tab3 as select * from (values (10, 100), (20, 200), (30, 300)) tab(m, k);

select
  t1.n,
  t2.m + t3.k as result,
  (select count(tab3.k) from tab3 where tab3.k = t1.n * 100 and tab3.m = t2.m) count
from 
  (values (1), (2), (3)) t1(n),
  tab2 t2 , tab3 t3
where t1.n = t2.n and t2.m = t3.m
order by result;

---

FIXED