Add Linux design philosophy optimization analysis for xAgent

[Copilot]

Comprehensive code review of the xAgent codebase analyzed through the lens of Unix/Linux design principles, documenting concrete, prioritized optimization recommendations.

Document: docs/linux_philosophy_optimization.md

Covers 10 Linux principles applied to the codebase with specific findings:

🔴 High Priority (correctness/security)

• Global logging.basicConfig() in agent.py — library code must not configure the root logger; breaks any host project's log setup
• CORS allow_origins=["*"] hardcoded in server implementation; should be policy-driven via config
• AgentInput has no upper bounds on max_iter, max_concurrent_tools, etc. — trivial DoS vector
• Silent background memory failures — retries exhaust, data is lost, caller never knows
• Filename typo: upstach_vector_store.py → upstash_vector_store.py

🟡 Medium Priority (maintainability/portability)

• agent.py (1,103 lines) and workflow.py (1,141 lines) violate single-responsibility; proposed split into focused submodules
• langfuse, chromadb, redis, upstash_vector should be optional extras, not hard dependencies
• Three-layer memory inheritance (Base → Basic → Local/Cloud) is unnecessarily deep; flatten with composition
• Magic numbers scattered across modules — centralize in xagent/defaults.py

🟢 Low Priority (UX/performance)

• Swarm class is a 22-line placeholder — implement or remove
• AutoWorkflow burns 6 LLM calls for trivial tasks; add a complexity fast-path
• Tool result preview capped at 20 chars — useless for debugging

Each finding includes a before/after code example and is mapped to its specific Linux principle (single responsibility, policy/mechanism separation, fail loudly, laziness, silence, portability, etc.). A three-phase implementation roadmap is included.

Original prompt

仔细阅读项目的所有代码，使用 Linux的设计哲学来审视这个项目，给出优化的完整建议，输出在markdown文件

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.