refactor: unify duplicate DAG construction (dag.py + ExecutionGraph)

[przemekboruta]

Summary

Closes #510

• Deletes dag.py and moves topologically_sort_column_configs into execution_graph.py as a module-level function
• Replaces networkx.topological_sort with an inline Kahn's algorithm, consistent with ExecutionGraph.get_topological_order
• Side-effect resolution is now O(1) via a side_effect_map dict — the previous implementation did a linear scan over sum(side_effect_dict.values(), []) which was O(n²)
• Updates imports in config_compiler.py and test_dag.py

Design note

The function is intentionally a module-level function, not a @classmethod on ExecutionGraph. ExecutionGraph is an execution abstraction (requires strategies, manages task scheduling); this function is a compilation step that works on raw ColumnConfigT without strategies. Mixing the two responsibilities would require either dummy strategies or a significant signature change to add_column.

Test plan

• Existing test_dag.py tests (test_dag_construction, test_circular_dependencies) cover the migrated function with updated import path
• ruff check passes on all modified files

[greptile-apps]

Greptile Summary

This PR consolidates two overlapping DAG implementations by deleting dag.py and moving topologically_sort_column_configs into execution_graph.py as a module-level function. The migration replaces the old networkx.topological_sort dependency with an inline Kahn's algorithm (consistent with ExecutionGraph.get_topological_order) and improves side-effect resolution from O(n²) linear scan to O(1) dict lookup. The Kahn's algorithm is correctly implemented, import paths in config_compiler.py and test_dag.py are updated, and no logic changes were introduced.

Confidence Score: 5/5

Safe to merge — clean mechanical refactor with no logic changes and correct Kahn's algorithm implementation.

All changes are a straightforward migration of a function between modules with an equivalent algorithm substitution. The Kahn's algorithm is correct, side-effect resolution is logically equivalent and more efficient, imports are updated in all call sites, and existing tests cover the migrated code. No P0 or P1 findings.

No files require special attention.

Vulnerabilities

No security concerns identified.

Important Files Changed

Filename	Overview
packages/data-designer-engine/src/data_designer/engine/dataset_builders/utils/execution_graph.py	Adds topologically_sort_column_configs as a module-level function using a correct inline Kahn's algorithm; the existing ExecutionGraph class is unchanged.
packages/data-designer-engine/src/data_designer/engine/dataset_builders/utils/config_compiler.py	Import updated from the deleted dag module to execution_graph; no logic changes.
packages/data-designer-engine/src/data_designer/engine/dataset_builders/utils/dag.py	File deleted as part of the consolidation; functionality migrated to execution_graph.py.
packages/data-designer-engine/tests/engine/dataset_builders/utils/test_dag.py	Import path updated to execution_graph; test logic is identical to the pre-PR version.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A["config_compiler.py\ncompile_dataset_builder_column_configs()"]
    B["execution_graph.py\ntopologically_sort_column_configs()"]
    C["ExecutionGraph.create()"]
    D["ExecutionGraph.get_topological_order()\n(Kahn's algorithm)"]
    E["DAGCircularDependencyError"]

    A -- "calls" --> B
    B -- "builds upstream/downstream\ndicts + Kahn's sort" --> D
    C -- "calls after building edges" --> D
    D -- "raises on cycle" --> E
    B -- "raises on cycle" --> E

    style B fill:#d4edda,stroke:#28a745
    style E fill:#f8d7da,stroke:#dc3545

Loading

_{Reviews (3): Last reviewed commit: "test: relax non-deterministic ordering a..." | Re-trigger Greptile}

[nabinchha]

@przemekboruta there's a larger PR in flight that touches these DAG abstractions. Let's wait until that merges before this one can re-base and merge!