feat(agent): add streamed reasoning summary callbacks to runAgent

readme.md

@@ -76,7 +76,7 @@ console.log(result.usage);   // { inputTokens, outputTokens, totalTokens, cacheR
 - ⚡ **Parallel tool execution** — multiple tool calls in a single step are executed concurrently
 - 🔧 **Auto-retry with backoff** — handles 429/529/503 errors with header-aware retry delays
 - 🩹 **Tool call repair** — case-insensitive name matching + invalid tool sink prevents crashes
-- 📊 **Token streaming** — `onToken` and `onToolCall` callbacks for real-time progress
+- 📊 **Token and reasoning streaming** — `onToken`, `onReasoning*`, and `onToolCall` callbacks for real-time progress
 - 💥 **Context overflow handling** — detects overflow and invokes your `onContextOverflow` callback
 
 ## Core API
@@ -98,6 +98,9 @@ The single entry point. Options:
 | `messages` | `ModelMessage[]` | `[]` | Conversation history (for multi-turn) |
 | `maxRetries` | `number` | `5` | Max retries on rate-limit/server errors |
 | `onToken` | `(delta: string) => void` | — | Streaming token callback |
+| `onReasoningStart` | `(id: string) => void` | — | Called when a reasoning summary starts |
+| `onReasoningDelta` | `(id: string, delta: string) => void` | — | Called for streamed reasoning summary text |
+| `onReasoningEnd` | `(id: string, text: string) => void` | — | Called when a reasoning summary completes |
 | `onToolCall` | `(name: string) => void` | — | Called when a tool is invoked |
 | `onToolResult` | `(name: string, result: unknown) => void` | — | Called when a tool finishes |
 | `validateCompletion` | `(result) => string \| void` | — | Return a string to reject and reprompt an incomplete run |