feat: add Ollama local provider support

- Added Ollama (local) as AI provider option
- Configure VPS endpoint for Ollama connection
- Auto-detect available models from Ollama server
- Support for vision-capable models (llama3.2-vision, etc)

types/providers.ts

@@ -1,4 +1,6 @@
-export type AIProvider = 'gemini' | 'openai' | 'anthropic' | 'azure';
+import { AIProvider } from './providers';
+
+export type AIProvider = 'gemini' | 'openai' | 'anthropic' | 'azure' | 'ollama';
 
 export interface ProviderConfig {
   id: AIProvider;
@@ -38,5 +40,13 @@ export const PROVIDERS: ProviderConfig[] = [
     models: ['gpt-4', 'gpt-4-32k', 'gpt-35-turbo'],
     requiresEndpoint: true,
     defaultModel: 'gpt-4'
+  },
+  {
+    id: 'ollama',
+    name: 'Ollama (Local)',
+    description: 'LLMs rodando localmente na sua VPS',
+    models: [],
+    requiresEndpoint: true,
+    defaultModel: 'llama3.2-vision'
   }
 ];