Milestone M2 — Direct Python vs PythonW Launch Validation

Issue: #5 (https://github.com/joelvaneenwyk/py-launch-lab/issues/5)

Description

Implement the process runner and validate the observable differences between launching scripts with python.exe (console) and pythonw.exe (GUI/no-console). This milestone produces hard evidence of console-window behaviour for each launch mode.

How it Works

1. runner.py spawns child processes with a controlled environment
2. The runner detects whether the spawned process opens a visible console window
3. It inspects the process tree for conhost.exe presence
4. Results record stdout_available, stderr_available, and visible_window_detected
5. Evidence artifacts are captured for each scenario

Tasks

• [ ] Implement runner.py: spawn child processes with controlled environment
• [ ] Detect whether spawned process opens a console window
• [ ] Detect conhost.exe appearing in process tree
• [ ] Record stdout_available, stderr_available, visible_window_detected
• [ ] Implement integration tests: test_python_vs_pythonw.py
• [ ] Capture evidence artifacts for python script.py vs pythonw script.py

Related Issues

• Depends on m1-static-pe-inspection.md - needs PE subsystem classification for result validation

Next Steps

• [ ] Run comparison tests on a Windows runner
• [ ] Verify evidence artifacts are written to artifacts/json/
• [ ] Confirm console-detection logic handles edge cases (e.g., redirected output)