feat(aidd-context): evaluate communication preferences

[alexsoyes]

🎯 What & why

Strengthens the project-memory AGENTS.md template with general, scannable communication preferences, then adds isolated scenarios to verify the resulting behavior without prescribing response templates.

🛠️ How it works

AGENTS.md consolidates first-line orientation, activation friction, visible causality, current-state tracking, and a single next move into one communication contract.

The new test route runs a selected scenario through a fresh subagent. The evaluator passes only AGENTS.md and the exact user prompt, then returns the response unedited for human judgment. This isolates the contract from conversation history and avoids encoding an expected answer.

Two communication scenarios cover decision pressure and a technical authentication regression. The plugin catalog is regenerated accordingly.

🧪 How to verify

• Invoke aidd-context:02-project-memory test.
• Select Authentication regression.
• Confirm the subagent response is returned unedited, followed by the human review checks.
• Confirm no result is marked accepted before the user gives a verdict.
• Run cd cli && pnpm test.

✅ I certify

• I DO CERTIFY I READ EACH LINE OF THE PULL REQUEST BECAUSE I AM A SOFTWARE ENGINEER, NOT A AI PUPPY.