Add testing infrastructure and refactor DAG transformation

Testing setup:
- Add pyproject.toml with mypy and pytest configuration
- Add requirements-dev.txt for development dependencies
- Create tests/ directory with test fixtures
- Add 17 unit tests for DAG transformation pipeline

Type annotations:
- Add app/types.py with TypedDict definitions for node configs
- Add typed helper functions: transform_node, build_input_name_mapping,
  bind_inputs, prepare_dag_for_execution
- Refactor run_recipe to use the new typed helpers

Regression tests for today's bugs:
- test_effect_cid_key_not_effect_hash: Verifies CID uses 'cid' key
- test_source_cid_binding_persists: Verifies bound CIDs in final DAG

Run tests with: pytest tests/ -v

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

app/types.py

@@ -0,0 +1,160 @@
+"""
+Type definitions for Art DAG L1 server.
+
+Uses TypedDict for configuration structures to enable mypy checking.
+"""
+
+from typing import Any, Dict, List, Optional, TypedDict, Union
+from typing_extensions import NotRequired
+
+
+# === Node Config Types ===
+
+class SourceConfig(TypedDict, total=False):
+    """Config for SOURCE nodes."""
+    cid: str  # Content ID (IPFS CID or SHA3-256 hash)
+    asset: str  # Asset name from registry
+    input: bool  # True if this is a variable input
+    name: str  # Human-readable name for variable inputs
+    description: str  # Description for variable inputs
+
+
+class EffectConfig(TypedDict, total=False):
+    """Config for EFFECT nodes."""
+    effect: str  # Effect name
+    cid: str  # Effect CID (for cached/IPFS effects)
+    # Effect parameters are additional keys
+    intensity: float
+    level: float
+
+
+class SequenceConfig(TypedDict, total=False):
+    """Config for SEQUENCE nodes."""
+    transition: Dict[str, Any]  # Transition config
+
+
+class SegmentConfig(TypedDict, total=False):
+    """Config for SEGMENT nodes."""
+    start: float
+    end: float
+    duration: float
+
+
+# Union of all config types
+NodeConfig = Union[SourceConfig, EffectConfig, SequenceConfig, SegmentConfig, Dict[str, Any]]
+
+
+# === Node Types ===
+
+class CompiledNode(TypedDict):
+    """Node as produced by the S-expression compiler."""
+    id: str
+    type: str  # "SOURCE", "EFFECT", "SEQUENCE", etc.
+    config: Dict[str, Any]
+    inputs: List[str]
+    name: NotRequired[str]
+
+
+class TransformedNode(TypedDict):
+    """Node after transformation for artdag execution."""
+    node_id: str
+    node_type: str
+    config: Dict[str, Any]
+    inputs: List[str]
+    name: NotRequired[str]
+
+
+# === DAG Types ===
+
+class CompiledDAG(TypedDict):
+    """DAG as produced by the S-expression compiler."""
+    nodes: List[CompiledNode]
+    output: str
+
+
+class TransformedDAG(TypedDict):
+    """DAG after transformation for artdag execution."""
+    nodes: Dict[str, TransformedNode]
+    output_id: str
+    metadata: NotRequired[Dict[str, Any]]
+
+
+# === Registry Types ===
+
+class AssetEntry(TypedDict, total=False):
+    """Asset in the recipe registry."""
+    cid: str
+    url: str
+
+
+class EffectEntry(TypedDict, total=False):
+    """Effect in the recipe registry."""
+    cid: str
+    url: str
+    temporal: bool
+
+
+class Registry(TypedDict):
+    """Recipe registry containing assets and effects."""
+    assets: Dict[str, AssetEntry]
+    effects: Dict[str, EffectEntry]
+
+
+# === Recipe Types ===
+
+class Recipe(TypedDict, total=False):
+    """Compiled recipe structure."""
+    name: str
+    version: str
+    description: str
+    owner: str
+    registry: Registry
+    dag: CompiledDAG
+    recipe_id: str
+    ipfs_cid: str
+    sexp: str
+    step_count: int
+    error: str
+
+
+# === API Request/Response Types ===
+
+class RecipeRunInputs(TypedDict):
+    """Mapping of input names to CIDs for recipe execution."""
+    # Keys are input names, values are CIDs
+    pass  # Actually just Dict[str, str]
+
+
+class RunResult(TypedDict, total=False):
+    """Result of a recipe run."""
+    run_id: str
+    status: str  # "pending", "running", "completed", "failed"
+    recipe: str
+    inputs: List[str]
+    output_cid: str
+    ipfs_cid: str
+    error: str
+    created_at: str
+    completed_at: str
+
+
+# === Helper functions for type narrowing ===
+
+def is_source_node(node: TransformedNode) -> bool:
+    """Check if node is a SOURCE node."""
+    return node.get("node_type") == "SOURCE"
+
+
+def is_effect_node(node: TransformedNode) -> bool:
+    """Check if node is an EFFECT node."""
+    return node.get("node_type") == "EFFECT"
+
+
+def is_variable_input(config: Dict[str, Any]) -> bool:
+    """Check if a SOURCE node config represents a variable input."""
+    return bool(config.get("input"))
+
+
+def get_effect_cid(config: Dict[str, Any]) -> Optional[str]:
+    """Get effect CID from config, checking both 'cid' and 'hash' keys."""
+    return config.get("cid") or config.get("hash")