docs: create ARCHITECTURE.md

[pmady]

Summary

Create an ARCHITECTURE.md documenting the project's design and data flow. Foundation reviewers and new contributors need this to understand the codebase quickly.

Suggested sections

1. Overview — what gpu-mcp-server does and where it fits
2. Data flow — NVML → Collector → Handler → MCP SDK → Transport → Agent
3. Key interfaces — Collector (gpu.go), Handler (server.go)
4. Build tags — nvml.go (cgo+linux), stub.go (everything else), mock.go (tests)
5. Transport modes — stdio (default), HTTP (optional)
6. Deployment — standalone binary, Docker, Kubernetes DaemonSet (Helm)
7. MIG support — how MIG instances are detected and reported

Reference

See keda-gpu-scaler's ARCHITECTURE.md for the expected level of detail.

Acceptance criteria

• ARCHITECTURE.md at repo root
• Covers all sections above
• Includes a text or Mermaid diagram of the data flow
• Linked from README

Difficulty: Beginner-Intermediate — documentation, requires reading the code