Maas api add gpu and restructure (#95)

* Working version with qwen

* Initial restructured commit

* Semi woking token based limiting, for enterprise for example it stops on 120 tokens but it should happen on 1000

* Restructuring as per PR comments + update token rate limiting, however there are still issues with tiers that needs to be debugged and fixed

* Fixes as per coderabbit comments

* fix: Remove duplicate HTTPRoute definition after rebase

- Removed duplicate gateway-routes.yaml from base
- HTTPRoute for maas-api is now only in base/networking/httproute.yaml
- Fixed kustomization build error after rebasing on main

* fix: Restore maas-api/deploy files that exist on main

- Restored kuadrant.yaml, gateway.yaml, httproute.yaml files
- Restored gateway-auth-policy.yaml for model policies
- Fixed duplicate ServiceAccount in deployment/samples/models
- All validations now pass except 2 ODH-related that also fail on main
  (external GitHub dependency issue with kustomize)

These files were incorrectly removed during rebase thinking they were
duplicates, but they serve different purposes than the ones in deployment/

* Fixes to deployment, added updated opendatahub/maas-api image, added a script for fast openshift deployment

* Referenced PR comments, tested deployment

* Improve scrit, removed dynamic AWS ELB address retrieving and added openshif route retrieving

* Removed kustomization config as it is redundant

* Updating policies

* Updating rbac and some minor changes around the install script and instructions

* Trying to move VLLMInferenceService to deployment/ not everything works, but simulator should work

* Updating instructions and cluster role issues

* Fixed qwen for LLMInferenceService, not finished facebook cpu model

* Removed conflicting parts from maas-api

* Revert maas-api/ directory to latest main branch state

---------

Co-authored-by: Jamie Land <hokie10@gmail.com> & Bartos

Author: dmytro-zaharnytskyi

.gitignore

@@ -37,4 +37,4 @@ pip-delete-this-directory.txt
 htmlcov/
 apps/frontend/.env.local
 apps/backend/.env
-CLAUDE.md
+CLAUDE.md

POLICY_METRICS_IMPLEMENTED.md

@@ -8,8 +8,8 @@ Our goal is to create a comprehensive platform for **Models as a Service** with
 ## 📦 Technology Stack
 
 - **Kuadrant/Authorino/Limitador**: API gateway and policy engine
-- **Istio**: Service mesh and traffic management
-- **Gateway API**: Traffic routing and management
+- **Gateway API**: Traffic routing and management (OpenShift native implementation)
+- **OpenShift Service Mesh**: Automatically provisioned when Gateway API is enabled (includes Istio)
 - **React**: Frontend framework
 - **Go**: Backend frameworks
 
@@ -26,7 +26,7 @@ Our goal is to create a comprehensive platform for **Models as a Service** with
 ## 🏗️ Architecture
 
 ### Backend Components
-- **API Gateway**: Istio/Envoy with Gateway API support and Kuadrant integration
+- **API Gateway**: OpenShift Gateway API implementation with Envoy proxy and Kuadrant integration
 - **Policy Engine**: Real-time policy enforcement through Kuadrant (Authorino + Limitador)
 - **Model Serving**: KServe-based AI model deployment with vLLM runtime
 - **Model Discovery**: Automatic model listing model resources
@@ -46,14 +46,25 @@ Our goal is to create a comprehensive platform for **Models as a Service** with
 
 ## 🚀 Quick Start
 
-For deployment instructions, see the READMEs in the deployment directory:
+### Deploy Infrastructure
 
-- **[Infrastructure](deployment/infrastructure/README.md)** - Base platform components (Istio, KServe, Kuadrant operators)
-- **[Example Usage](deployment/examples/README.md)** - Complete deployment examples with models, authentication, and observability
+See the comprehensive [Deployment Guide](deployment/README.md) for detailed instructions.
 
-## Development Setup
+Quick deployment for OpenShift:
+```bash
+export CLUSTER_DOMAIN="apps.your-openshift-cluster.com"
+kustomize build deployment/overlays/openshift | envsubst | kubectl apply -f -
+```
+
+Quick deployment for Kubernetes:
+```bash
+export CLUSTER_DOMAIN="your-kubernetes-domain.com"
+kustomize build deployment/overlays/kubernetes | envsubst | kubectl apply -f -
+```
+
+### Start Development Environment
 
-After deploying the infrastructure, start the frontend and backend:
+After deploying the infrastructure:
 
 #### Option A: One-Command Start (Recommended)
 ```bash
@@ -107,117 +118,94 @@ This will:
 ## 🔧 Development
 
 ### Project Structure
-```
-maas-billing/
-├── apps/
-│   ├── frontend/          # React frontend with Material-UI
-│   │   ├── src/
-│   │   │   ├── components/    # Policy Manager, Metrics Dashboard, etc.
-│   │   │   ├── hooks/         # API integration hooks
-│   │   │   └── services/      # API client
-│   │   └── package.json
-│   └── backend/           # Node.js/Express API server
-│       ├── src/
-│       │   ├── routes/        # API endpoints
-│       │   ├── services/      # Kuadrant integration
-│       │   └── utils/         # Logging and utilities
-│       └── package.json
-├── deployment/kuadrant/   # Kuadrant infrastructure
-└── start-*.sh           # Development scripts
-```
 
-### API Endpoints
-- `GET /api/v1/policies` - List all policies
-- `POST /api/v1/policies` - Create new policy
-- `PUT /api/v1/policies/:id` - Update policy
-- `DELETE /api/v1/policies/:id` - Delete policy
-- `GET /api/v1/metrics/live-requests` - Real-time metrics
-- `GET /api/v1/metrics/dashboard` - Dashboard statistics
+| Directory | Description | Documentation |
+|-----------|-------------|---------------|
+| `apps/frontend/` | React frontend with Material-UI | [Frontend Guide](apps/frontend/README.md) |
+| `apps/backend/` | Node.js/Express API server | [Backend Guide](apps/backend/README.md) |
+| `maas-api/` | Go API for key management | [MaaS API Guide](maas-api/README.md) |
+| `deployment/` | Kubernetes/OpenShift deployments | [Deployment Guide](deployment/README.md) |
+| `scripts/` | Automation and utility scripts | - |
 
-### Environment Variables
-```bash
-# Backend (.env)
-PORT=3001
-FRONTEND_URL=http://localhost:3000
-```
+### Available Scripts
 
-## 🛑 Stopping the Platform
+From the repository root:
+- `./start-dev.sh` - Start full development environment
+- `./stop-dev.sh` - Stop all development services
+- `./start-backend.sh` - Start backend only
+- `./start-frontend.sh` - Start frontend only
+- `./scripts/test-gateway.sh` - Test gateway endpoints
 
-```bash
-# Stop all services
-./stop-dev.sh
+### Backend API Endpoints
 
-# Or manually stop individual components
-pkill -f "npm start"    # Stop frontend
-pkill -f "npm run dev"  # Stop backend
-```
+The backend provides these key endpoints:
+- `GET /api/v1/models` - List available models
+- `GET /api/v1/policies` - Retrieve current policies
+- `POST /api/v1/policies` - Create/update policies
+- `GET /api/v1/metrics/live-requests` - Live metrics stream
+- `POST /api/v1/simulator/run` - Run policy simulation
 
-## 📊 Monitoring & Logs
+### Frontend Components
 
-### Application Logs
-```bash
-# Real-time logs
-tail -f backend.log     # Backend API logs
-tail -f frontend.log    # Frontend build logs
-
-# Service logs
-kubectl logs -n kuadrant-system -l app=limitador
-kubectl logs -n kuadrant-system -l app=authorino
-```
+Key React components:
+- `PolicyBuilder` - Drag-and-drop policy editor
+- `MetricsDashboard` - Real-time metrics visualization
+- `RequestSimulator` - Policy testing interface
+- `TokenManagement` - API key management
 
-### Metrics and Health Checks
-- Backend health: `curl http://localhost:3001/health`
-- Kuadrant status: `kubectl get pods -n kuadrant-system`
-- Live metrics: `curl http://localhost:3001/api/v1/metrics/live-requests`
+## 🧪 Testing
 
-## 🔍 Troubleshooting
-
-### Common Issues
-
-**Port Already in Use**
+### Test Infrastructure
 ```bash
-# Kill processes on ports 3000/3001
-lsof -ti:3000 | xargs kill -9
-lsof -ti:3001 | xargs kill -9
-```
+# Use the test script
+./scripts/test-gateway.sh
 
-**Kuadrant Not Ready**
-```bash
-# Check Kuadrant deployment
-kubectl get pods -n kuadrant-system
-kubectl get gateways -A
+# Or manually test endpoints
+curl http://localhost:3001/health
 ```
 
-**Frontend Not Loading**
+### Run Frontend Tests
 ```bash
-# Clear browser cache and restart frontend
-rm -rf apps/frontend/node_modules/.cache
-./start-frontend.sh
+cd apps/frontend
+npm test
 ```
 
-**No Metrics Data**
+### Run Backend Tests
 ```bash
-# Check Kuadrant components
-kubectl port-forward -n kuadrant-system svc/limitador 8080:8080
-curl http://localhost:8080/metrics
+cd apps/backend
+npm test
 ```
 
+## 📚 Documentation
+
+- [Deployment Guide](deployment/README.md) - Complete deployment instructions
+- [Platform-Specific Overlays](deployment/overlays/README.md) - OpenShift vs Kubernetes
+- [MaaS API Documentation](maas-api/README.md) - Go API for key management
+- [OAuth Setup Guide](OAUTH_SETUP.md) - Configure OAuth authentication
+
 ## 🤝 Contributing
 
+We welcome contributions! Please:
 1. Fork the repository
-2. Create a feature branch: `git checkout -b feature/amazing-feature`
-3. Commit changes: `git commit -m 'Add amazing feature'`
-4. Push to branch: `git push origin feature/amazing-feature`
-5. Open a Pull Request
+2. Create a feature branch
+3. Make your changes
+4. Submit a pull request
+
+## 📝 License
 
-## 📄 License
+This project is licensed under the Apache 2.0 License.
 
-This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
+## 🙏 Acknowledgments
 
----
+Built with:
+- [Kuadrant](https://kuadrant.io/) for API management
+- [KServe](https://kserve.github.io/) for model serving
+- [OpenShift](https://www.openshift.com/) with Service Mesh for infrastructure
+- [React](https://react.dev/) and [Material-UI](https://mui.com/)
 
-## 📚 Additional Resources
+## 📞 Support
 
-- **Kuadrant Documentation**: https://kuadrant.io/
-- **KServe Documentation**: https://kserve.github.io/website/
-- **Istio Gateway API**: https://istio.io/latest/docs/tasks/traffic-management/ingress/gateway-api/
+For questions or issues:
+- Open an issue on GitHub
+- Check the [deployment guide](deployment/README.md) for troubleshooting
+- Review the [samples](deployment/samples/models/) for examples