feat(openai): add getModelSetup() and typed provider options for OpenAI reasoning settings

readme.md

@@ -70,11 +70,43 @@ const options: ISmartAiOptions = {
   baseUrl: 'http://localhost:11434',
   // Ollama-only: model runtime options
   ollamaOptions: { think: true, num_ctx: 4096 },
+  // OpenAI request-time options for generateText()/streamText()
+  providerOptions: { openai: { reasoningEffort: 'xhigh' } },
 };
 
 const model = getModel(options);
 ```
 
+### `getModelSetup(options): { model, providerOptions }`
+
+Use this when a provider needs request-time AI SDK `providerOptions`, such as GPT-5 reasoning settings for OpenAI. `getModel()` remains available for existing consumers; `getModelSetup()` returns the same model plus provider options to pass into `generateText()` or `streamText()`.
+
+```typescript
+import { getModelSetup, generateText } from '@push.rocks/smartai';
+
+const setup = getModelSetup({
+  provider: 'openai',
+  apiKey: process.env.OPENAI_API_KEY,
+  model: 'gpt-5.5',
+  providerOptions: {
+    openai: {
+      reasoningEffort: 'xhigh',
+      textVerbosity: 'high',
+    },
+  },
+});
+
+const result = await generateText({
+  model: setup.model,
+  prompt: 'Draft a careful migration plan.',
+  providerOptions: setup.providerOptions,
+});
+
+console.log(result.text);
+```
+
+OpenAI `reasoningEffort` supports `'none'`, `'minimal'`, `'low'`, `'medium'`, `'high'`, and `'xhigh'`. Model IDs are accepted as strings, so new IDs like `'gpt-5.5'` can be used before upstream model unions are updated.
+
 ### Re-exported AI SDK Functions
 
 SmartAI re-exports the most commonly used functions from `ai` for convenience:
@@ -101,7 +133,7 @@ import type {
 | Provider | Package | Example Models |
 |----------|---------|----------------|
 | **Anthropic** | `@ai-sdk/anthropic` | `claude-sonnet-4-5-20250929`, `claude-opus-4-5-20250929` |
-| **OpenAI** | `@ai-sdk/openai` | `gpt-4o`, `gpt-4o-mini`, `o3-mini` |
+| **OpenAI** | `@ai-sdk/openai` | `gpt-5.5`, `gpt-5`, `gpt-4o`, `o3-mini` |
 | **Google** | `@ai-sdk/google` | `gemini-2.0-flash`, `gemini-2.5-pro` |
 | **Groq** | `@ai-sdk/groq` | `llama-3.3-70b-versatile`, `mixtral-8x7b-32768` |
 | **Mistral** | `@ai-sdk/mistral` | `mistral-large-latest`, `mistral-small-latest` |