refactor: unify SQL planning for ORDER BY, HAVING, DISTINCT, etc

[lichuang]

Which issue does this PR close?

This PR unifies SQL planning logic for ORDER BY with HAVING and other clauses by adopting the merged schema approach (#10326).

Problem: DataFusion previously had two different code paths for handling ORDER BY:

1. add_missing_columns - traverses the plan tree to find Projection nodes and injects missing columns
2. Merged Schema approach (used by HAVING) - uses a unified schema to resolve expressions directly

This duality caused complex, hard-to-maintain code, non-intuitive handling of queries like SELECT x FROM foo ORDER BY y,and unnecessarily wrapped execution plans.

Solution: Refactor ORDER BY planning to use the merged schema approach, directly adding missing expressions to the SELECT list instead of traversing the plan tree.

Key Changes:
• Simplified order_by_to_sort_expr: Removed the additional_schema parameter and the internal schema merging logic - now input_schema directly instead of building a combined schema
• Added add_missing_order_by_exprs(): A new function that handles missing ORDER BY expressions by adding them directly the SELECT list before planning
• Properly handles aggregate/window functions, column references, and aliases
• Returns error for SELECT DISTINCT with missing ORDER BY expressions
• Supports strict mode (when GROUP BY is present) to preserve proper error messages
• Unified planning flow: ORDER BY expressions are now resolved against the SELECT list first, then missing ones are adde eliminating the need for complex plan tree traversal (add_missing_columns)

This makes the code cleaner, more maintainable, and produces simpler execution plans.

Rationale for this change

What changes are included in this PR?

Are these changes tested?

Are there any user-facing changes?

[lichuang]

@alamb @jonahgao

[alamb]

This seems to add more code (rather than unify the code paths as the PR comment suggests) 🤔

[alamb]

In other words, this PR seems to add more code paths, when the idea was to reduce the code / duplication

[lichuang]

@alamb thanks comment. After comment, I Simplified order_by_to_sort_expr by removeing the additional_schema parameter, which is introduced in pr #10234.