test: Add comprehensive workflow integration tests (195 tests)

Add integration test suites for workflow modules with mock LLM fixtures:

- test_document_gen_workflow.py: 54 tests for DocumentGenerationWorkflow
  (token costs, chunked generation, export functionality)
- test_refactor_plan_workflow.py: 40 tests for RefactorPlanWorkflow
  (tech debt scanning, trajectory analysis, priority scoring)
- test_perf_audit_workflow.py: 39 tests for PerformanceAuditWorkflow
  (anti-pattern detection, complexity analysis, hotspot identification)
- test_step_config.py: 41 tests for WorkflowStepConfig
  (creation, effective tier, overrides, serialization, validation)
- test_workflow_base.py: Extended with ~45 tests for BaseWorkflow
  (initialization, execution, progress tracking, XML prompts)
- conftest.py: Added mock fixtures (mock_llm_responses, mock_anthropic_client)

Coverage improvements:
- step_config.py: 36% → 100%
- base.py: 36% → 88%
- refactor_plan.py: 7% → 89%
- perf_audit.py: 8% → 86%
- document_gen.py: 6% → 79%

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <[email protected]>

Author: 2 people

tests/conftest.py

@@ -121,3 +121,187 @@ def workflow_config():
         "enable_xml_prompts": False,
         "skip_expensive_stages": True,
     }
+
+
+# =============================================================================
+# Integration Test Fixtures - Mock LLM Responses
+# =============================================================================
+
+
+@pytest.fixture
+def mock_llm_responses():
+    """
+    Configurable mock LLM responses for integration testing.
+
+    Usage:
+        def test_something(mock_llm_responses):
+            mock_llm_responses["security_scan"] = "No vulnerabilities found"
+            # Run workflow...
+    """
+    return {
+        # Code review responses
+        "classify": "feature: Adding new functionality",
+        "security_scan": '{"vulnerabilities": [], "score": 100}',
+        "code_analysis": '{"quality_score": 85, "issues": []}',
+        "architect_review": "Architecture looks good. No concerns.",
+        # Security audit responses
+        "identify_vulnerabilities": '{"vulnerabilities": [], "risk_level": "low"}',
+        "analyze_risk": '{"overall_risk": "low", "recommendations": []}',
+        "generate_report": "# Security Audit Report\n\nNo issues found.",
+        # Document generation responses
+        "analyze_structure": '{"modules": [], "functions": []}',
+        "generate_documentation": "# API Documentation\n\nGenerated docs here.",
+        # Health check responses
+        "diagnose": '{"health_score": 95, "issues": []}',
+        "fix": '{"fixes_applied": 0, "fixes": []}',
+        # Generic fallback
+        "default": "Task completed successfully.",
+    }
+
+
+@pytest.fixture
+def mock_anthropic_client(mock_llm_responses):
+    """
+    Mock Anthropic client for integration tests.
+
+    Returns deterministic responses based on the prompt content.
+    """
+    from unittest.mock import MagicMock
+
+    def create_mock_response(prompt: str) -> MagicMock:
+        """Create a mock response based on prompt content."""
+        response = MagicMock()
+
+        # Determine response based on keywords in prompt
+        content = mock_llm_responses["default"]
+        for key, value in mock_llm_responses.items():
+            if key.lower() in prompt.lower():
+                content = value
+                break
+
+        response.content = [MagicMock(text=content)]
+        response.usage = MagicMock(input_tokens=100, output_tokens=50)
+        return response
+
+    client = MagicMock()
+    client.messages.create = MagicMock(
+        side_effect=lambda **kwargs: create_mock_response(
+            kwargs.get("messages", [{}])[-1].get("content", "")
+            if isinstance(kwargs.get("messages", [{}])[-1], dict)
+            else str(kwargs.get("messages", []))
+        )
+    )
+
+    return client
+
+
+@pytest.fixture
+def patch_llm_client(mock_anthropic_client):
+    """
+    Patch the Anthropic client globally for integration tests.
+
+    Usage:
+        def test_workflow(patch_llm_client):
+            # LLM calls will return mock responses
+            result = await workflow.execute(...)
+    """
+    from unittest.mock import patch
+
+    with patch("anthropic.Anthropic", return_value=mock_anthropic_client):
+        yield mock_anthropic_client
+
+
+@pytest.fixture
+def sample_python_file(temp_dir):
+    """Create a sample Python file for testing file-based workflows."""
+    code = '''"""Sample module for testing."""
+
+def add(a: int, b: int) -> int:
+    """Add two numbers."""
+    return a + b
+
+def divide(a: int, b: int) -> float:
+    """Divide a by b."""
+    if b == 0:
+        raise ValueError("Cannot divide by zero")
+    return a / b
+
+class Calculator:
+    """Simple calculator class."""
+
+    def __init__(self):
+        self.history = []
+
+    def calculate(self, op: str, a: int, b: int) -> float:
+        if op == "add":
+            result = add(a, b)
+        elif op == "divide":
+            result = divide(a, b)
+        else:
+            raise ValueError(f"Unknown operation: {op}")
+        self.history.append((op, a, b, result))
+        return result
+'''
+    file_path = temp_dir / "sample.py"
+    file_path.write_text(code)
+    return file_path
+
+
+@pytest.fixture
+def sample_project_dir(temp_dir):
+    """Create a sample project directory structure for testing."""
+    # Create src directory
+    src_dir = temp_dir / "src"
+    src_dir.mkdir()
+
+    # Create main module
+    (src_dir / "__init__.py").write_text("")
+    (src_dir / "main.py").write_text(
+        '''
+"""Main application module."""
+
+def main():
+    print("Hello, World!")
+
+if __name__ == "__main__":
+    main()
+'''
+    )
+
+    # Create utils module
+    (src_dir / "utils.py").write_text(
+        '''
+"""Utility functions."""
+
+def format_name(first: str, last: str) -> str:
+    """Format a full name."""
+    return f"{first} {last}"
+'''
+    )
+
+    # Create tests directory
+    tests_dir = temp_dir / "tests"
+    tests_dir.mkdir()
+    (tests_dir / "__init__.py").write_text("")
+    (tests_dir / "test_main.py").write_text(
+        '''
+"""Tests for main module."""
+
+def test_placeholder():
+    assert True
+'''
+    )
+
+    # Create pyproject.toml
+    (temp_dir / "pyproject.toml").write_text(
+        """
+[project]
+name = "sample-project"
+version = "0.1.0"
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+"""
+    )
+
+    return temp_dir