docs: Add comprehensive bilingual documentation with MkDocs site

[tianhao909]

Summary

This PR adds comprehensive, self-hosted bilingual (English/Chinese) documentation for kvcached using MkDocs Material, complementing the existing DeepWiki integration.

Motivation: Why local docs in addition to DeepWiki?

Dimension	DeepWiki (current)	This PR (new)
Reliability	Third-party service, no SLA	In-repo, always available
Version sync	AI-generated, may be outdated	Same repo as code, updated via PRs
Community contribution	Cannot edit directly	Anyone can submit doc PRs
Offline access	Requires internet	mkdocs serve works locally
Multi-language	English only	English + Chinese (i18n)
Content depth	AI summary level	Paper-informed technical docs
Structure	Auto-generated	User-journey organized
SEO/Search	DeepWiki platform only	Independent site with full-text search

Note: DeepWiki is preserved as an additional resource in the README.

What's Included

New files:

• mkdocs.yml — MkDocs Material configuration with i18n
• requirements-docs.txt — Documentation build dependencies
• README_cn.md — Full Chinese README
• docs/en/ — 23 English documentation pages
• docs/zh/ — 23 Chinese documentation pages

Documentation structure:

• Getting Started: Installation, Quick Start, Docker
• Core Concepts: Architecture, GPU Virtual Memory, Autopatch, Prefix Caching
• User Guide: Multi-model, Memory Control, Router/Sleep, Inference+Finetune, Serverless, Colocation
• Configuration: Environment Variables, Controller, Engine Options
• Performance: Benchmarks, Tuning
• Development: Contributing, Build System, Testing
• FAQ: Common questions and troubleshooting

Modified files:

• README.md — Updated Documentation section with local docs links, added language switcher

Technical highlights

• Architecture docs incorporate insights from the Prism (OSDI 2026) and GPU Multitasking papers
• Mermaid diagrams for system architecture and data flow
• Code examples for all major use cases
• Complete environment variable reference
• Bilingual support with MkDocs i18n plugin

How to preview

pip install -r requirements-docs.txt
mkdocs serve
# Open http://127.0.0.1:8000

Checklist

• English documentation (23 pages)
• Chinese documentation (23 pages)
• MkDocs Material configuration
• README.md updated with doc links
• README_cn.md added
• No existing code modified

[Copilot]

Pull request overview

This PR introduces an in-repo, self-hostable MkDocs (Material) documentation site for kvcached, adding bilingual (English/Chinese) docs and updating the root README to point users to the local docs (while keeping DeepWiki as an additional resource).

Changes:

• Add MkDocs Material configuration (with mkdocs-static-i18n) and a dedicated docs requirements file.
• Add a full English documentation set under docs/en/ and a Chinese documentation set under docs/zh/, plus a Chinese top-level README.
• Update README.md to include local-docs build/serve instructions and links into the docs tree.

Reviewed changes

Copilot reviewed 50 out of 50 changed files in this pull request and generated 21 comments.

Show a summary per file

File	Description
requirements-docs.txt	Adds Python dependencies for building/serving the MkDocs site locally.
README.md	Adds language switcher and updates documentation section to reference local MkDocs docs.
README_cn.md	Adds a Chinese README (currently duplicated content needs cleanup).
mkdocs.yml	Adds MkDocs Material configuration, navigation, and i18n plugin setup.
docs/zh/index.md	Adds Chinese docs landing page.
docs/zh/getting-started/installation.md	Adds Chinese installation guide.
docs/zh/getting-started/quick-start.md	Adds Chinese quick start guide.
docs/zh/getting-started/docker.md	Adds Chinese Docker page (currently contains literal \n escapes).
docs/zh/core-concepts/architecture.md	Adds Chinese architecture overview.
docs/zh/core-concepts/virtual-memory.md	Adds Chinese virtual memory page (currently contains literal \n escapes).
docs/zh/core-concepts/autopatch.md	Adds Chinese autopatch page (currently contains literal \n escapes).
docs/zh/core-concepts/prefix-caching.md	Adds Chinese prefix caching page (currently contains literal \n escapes).
docs/zh/user-guide/multi-model.md	Adds Chinese multi-model guide (currently contains literal \n escapes).
docs/zh/user-guide/memory-control.md	Adds Chinese memory-control guide (currently contains literal \n escapes).
docs/zh/user-guide/router-sleep.md	Adds Chinese router/sleep guide (currently contains literal \n escapes).
docs/zh/user-guide/serverless.md	Adds Chinese serverless page (currently contains literal \n escapes).
docs/zh/user-guide/inference-finetune.md	Adds Chinese inference+finetune page (currently contains literal \n escapes).
docs/zh/user-guide/workload-colocation.md	Adds Chinese colocation page (currently contains literal \n escapes).
docs/zh/configuration/environment.md	Adds Chinese env var page (currently contains literal \n escapes).
docs/zh/configuration/controller.md	Adds Chinese controller config page (currently contains literal \n escapes).
docs/zh/configuration/engine-options.md	Adds Chinese engine options page (currently contains literal \n escapes).
docs/zh/performance/benchmarks.md	Adds Chinese benchmarks page (currently contains literal \n escapes).
docs/zh/performance/tuning.md	Adds Chinese tuning page (currently contains literal \n escapes).
docs/zh/development/contributing.md	Adds Chinese contributing page (currently contains literal \n escapes).
docs/zh/development/build-system.md	Adds Chinese build-system page (currently contains literal \n escapes).
docs/zh/development/testing.md	Adds Chinese testing page (currently contains literal \n escapes).
docs/zh/faq.md	Adds Chinese FAQ page.
docs/en/index.md	Adds English docs landing page.
docs/en/getting-started/installation.md	Adds English installation guide.
docs/en/getting-started/quick-start.md	Adds English quick start guide (includes Material tabbed sections).
docs/en/getting-started/docker.md	Adds English Docker deployment guide.
docs/en/core-concepts/architecture.md	Adds English architecture documentation (with Mermaid diagrams).
docs/en/core-concepts/virtual-memory.md	Adds English deep-dive on CUDA/HIP VMM-based virtual memory.
docs/en/core-concepts/autopatch.md	Adds English autopatch system documentation.
docs/en/core-concepts/prefix-caching.md	Adds English prefix caching documentation.
docs/en/user-guide/multi-model.md	Adds English multi-model serving guide.
docs/en/user-guide/memory-control.md	Adds English memory-control CLI guide.
docs/en/user-guide/router-sleep.md	Adds English router/sleep management guide.
docs/en/user-guide/serverless.md	Adds English serverless guide.
docs/en/user-guide/inference-finetune.md	Adds English inference+finetune colocation guide.
docs/en/user-guide/workload-colocation.md	Adds English workload colocation guide.
docs/en/configuration/environment.md	Adds English environment variable reference.
docs/en/configuration/controller.md	Adds English controller configuration guide.
docs/en/configuration/engine-options.md	Adds English engine options guide (copy/paste fixes needed for shell line continuations).
docs/en/performance/benchmarks.md	Adds English benchmarks overview and how to run benchmarks.
docs/en/performance/tuning.md	Adds English tuning guide.
docs/en/development/contributing.md	Adds English contributing guide (tooling + SPDX header guidance).
docs/en/development/build-system.md	Adds English build-system overview.
docs/en/development/testing.md	Adds English testing guide.
docs/en/faq.md	Adds English FAQ page.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.