feat: add llm-katan to k8s deploy

[JaredforReal]

What this PR does / why we need it:

• add llm-katan to k8s deploy
• separate core and llm-katan mode, as what we do in Docker Compose deploy
• update README and docs
• expand pvc size
• fix inference-pool selector error

[netlify]

✅ Deploy Preview for vllm-semantic-router ready!

Name	Link
🔨 Latest commit	ee082b4
🔍 Latest deploy log	https://app.netlify.com/projects/vllm-semantic-router/deploys/68f5deff9dc72700088a047b
😎 Deploy Preview	https://deploy-preview-466--vllm-semantic-router.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[github-actions]

👥 vLLM Semantic Team Notification

The following members have been identified for the changed files in this PR and have been automatically assigned:

📁 deploy

Owners: @rootfs, @Xunzhuo
Files changed:

• deploy/kubernetes/base/config.yaml
• deploy/kubernetes/base/deployment.yaml
• deploy/kubernetes/base/kustomization.yaml
• deploy/kubernetes/base/pv.yaml
• deploy/kubernetes/base/tools_db.json
• deploy/kubernetes/overlays/core/kustomization.yaml
• deploy/kubernetes/overlays/llm-katan/kustomization.yaml
• deploy/kubernetes/overlays/llm-katan/patch-llm-katan.yaml
• deploy/kubernetes/overlays/storage/kustomization.yaml
• deploy/kubernetes/overlays/storage/namespace.yaml
• deploy/kubernetes/README.md
• deploy/kubernetes/ai-gateway/inference-pool/inference-pool.yaml
• deploy/kubernetes/kustomization.yaml
• deploy/kubernetes/base/namespace.yaml
• deploy/kubernetes/base/service.yaml
• deploy/kubernetes/overlays/storage/pvc.yaml

📁 website

Owners: @Xunzhuo, @rootfs, @yuluo-yx
Files changed:

• website/docs/installation/kubernetes.md
• website/docs/troubleshooting/network-tips.md

---

🎉 Thanks for your contributions!

This comment was automatically generated based on the OWNER files in the repository.

[yossiovadia]

look solid for moving forward, but consider adding validation for the Qwen model download and possibly making the init container more robust in handling download failures, but it can be a future enhancement.