refactor: extract provider logic into modular registry

Each AI provider now has its own module implementing a ProviderModule
interface with createGateway, getAuthHeaders, and fetchModels methods.
This replaces the monolithic switch statements in ai-provider.ts and
models.post.ts with a clean registry pattern.

app/types/model.ts

@@ -1,30 +1,5 @@
 import * as schema from '~~/drizzle/schema';
 
-export const Providers = [
-    'openrouter',
-    'ollama',
-    'vllm',
-    'cerebras',
-    'google',
-    'longcat',
-    'cohere',
-    'inception',
-    'mistral',
-    'closedrouter',
-] as const;
 export const SupportedModalities = ['text', 'image', 'audio', 'video', 'pdf'] as const;
 
-export const providerBaseUrls = {
-    openrouter: 'https://openrouter.ai/api/v1',
-    vllm: '',
-    ollama: '',
-    cerebras: 'https://api.cerebras.ai/v1',
-    google: 'https://generativelanguage.googleapis.com/v1beta',
-    longcat: 'https://api.longcat.chat/openai/v1',
-    cohere: 'https://api.cohere.ai/v2',
-    inception: 'https://api.inceptionlabs.ai/v1',
-    mistral: 'https://api.mistral.ai/v1',
-    closedrouter: 'https://router.queef.in/v1',
-};
-
 export type Model = typeof schema.models.$inferSelect;