fix(runner): seed clean workspace task brief

[YOMXXX]

Summary

• seed the default CleanContextRunner workspace with compact AGENTS.md and SOUL.md task briefs so OpenClaw lightweight runs do not inject missing 76-byte stubs
• preserve the explicit system prompt/user prompt path and disabled-tools behavior for L1 extraction and other text-only runner calls
• add a regression test that clears the default clean workspace and verifies the brief files before the embedded agent receives the run

Verification

• npm test
• npm run build

Fixes #176
Related: #179

[Maxwell-Code07]

Awesome work! 13 fixes in one batch covering Gateway hardening, embedding compatibility, Hermes memory injection, offload stability, and pipeline optimization — a very productive cleanup session 👏

We'll review and merge them one by one 🚀