Initial commit: LlamaLink Go rewrite

Complete rewrite from Python/FastAPI to Go/Gin:
- Go backend: auth (API keys + bcrypt), llama.cpp subprocess manager,
  hot-swap multi-model, rate limiting, quota system, webhooks
- Vue 3 SPA admin panel (src/) with Tailwind CSS
- Deployment: Docker multi-stage, docker-compose, nginx, systemd
- GORM/SQLite models: ApiKey, Model, UsageLog, Quota, Webhook
- REST API: /api/v1/admin/* (keys, models, chat, usage, health)
- Embedded frontend via go:embed (build output at web/dist/)

Removed legacy Python artifacts (app/, tests/, pyproject.toml, etc.)

internal/api/router.go

@@ -0,0 +1,77 @@
+package api
+
+import (
+	"github.com/gin-gonic/gin"
+	"github.com/llamalink/llamalink/internal/api/handlers"
+	"github.com/llamalink/llamalink/internal/api/middleware"
+	"github.com/llamalink/llamalink/internal/auth"
+	"github.com/llamalink/llamalink/internal/config"
+	"github.com/llamalink/llamalink/internal/llama"
+	"github.com/llamalink/llamalink/internal/quota"
+	"gorm.io/gorm"
+)
+
+func New(cfg *config.Config, db *gorm.DB, llamaManager *llama.Manager) *gin.Engine {
+	if cfg.LlamalinkEnv == "production" {
+		gin.SetMode(gin.ReleaseMode)
+	}
+
+	r := gin.New()
+	r.Use(gin.Recovery())
+	r.Use(gin.Logger())
+
+	// Initialize services
+	authService := auth.NewService(db)
+	quotaSvc := quota.NewService(db)
+	webhookSvc := quota.NewWebhookService(db)
+	proxy := llama.NewProxy(llamaManager)
+
+	// Initialize handlers
+	healthHandler := handlers.NewHealthHandler(db, llamaManager)
+	chatHandler := handlers.NewChatHandler(proxy, authService, quotaSvc, webhookSvc)
+	keysHandler := handlers.NewKeysHandler(authService)
+	modelsHandler := handlers.NewModelsHandler(llamaManager)
+	usageHandler := handlers.NewUsageHandler(db, quotaSvc)
+
+	// Health endpoints (public)
+	r.GET("/health", healthHandler.Health)
+	r.GET("/ready", healthHandler.Ready)
+
+	// API v1 group
+	v1 := r.Group("/v1")
+
+	// Chat completions (requires API key auth + chat scope)
+	chat := v1.Group("/chat")
+	chat.Use(middleware.APIKeyAuth(authService))
+	chat.POST("/completions", chatHandler.ChatCompletions)
+
+	// Keys management (admin only)
+	keys := v1.Group("/keys")
+	keys.Use(middleware.APIKeyAuth(authService))
+	keys.Use(middleware.AdminOnly())
+	keys.GET("", keysHandler.ListKeys)
+	keys.POST("", keysHandler.CreateKey)
+	keys.DELETE("/:id", keysHandler.RevokeKey)
+
+	// Models management
+	models := v1.Group("/models")
+	models.Use(middleware.APIKeyAuth(authService))
+	models.GET("", modelsHandler.ListModels)
+	models.POST("", modelsHandler.CreateModel)
+	models.GET("/active", modelsHandler.GetActiveModel)
+	models.GET("/:name", modelsHandler.GetActiveModel) // alias for compatibility
+
+	// Model load (admin only)
+	modelLoad := v1.Group("/models")
+	modelLoad.Use(middleware.APIKeyAuth(authService))
+	modelLoad.Use(middleware.AdminOnly())
+	modelLoad.POST("/:name/load", modelsHandler.LoadModel)
+
+	// Usage (requires API key auth)
+	usage := v1.Group("/usage")
+	usage.Use(middleware.APIKeyAuth(authService))
+	usage.GET("", usageHandler.GetCurrentKeyUsage)
+	usage.GET("/:key_id", usageHandler.GetUsage)
+
+	return r
+}