refactor(ai): Add completion signal to onText/onThinking callbacks

- Update LLMOptions interface to include completion boolean parameter
- Modify all providers to signal when text/thinking blocks are complete
- Update examples to handle the completion parameter
- Move documentation files to docs/ directory

packages/ai/test/examples/anthropic.ts

@@ -24,8 +24,8 @@ const tools: Tool[] = [
 ];
 
 const options: AnthropicLLMOptions = {
-    onText: (t) => process.stdout.write(t),
-    onThinking: (t) => process.stdout.write(chalk.dim(t)),
+    onText: (t, complete) => process.stdout.write(t + (complete ? "\n" : "")),
+    onThinking: (t, complete) => process.stdout.write(chalk.dim(t + (complete ? "\n" : ""))),
     thinking: { enabled: true }
 };
 const ai = new AnthropicLLM("claude-sonnet-4-0", process.env.ANTHROPIC_OAUTH_TOKEN ?? process.env.ANTHROPIC_API_KEY);

packages/ai/test/examples/openai-completions.ts

@@ -21,8 +21,8 @@ const tools: Tool[] = [
 ];
 
 const options: OpenAICompletionsLLMOptions = {
-    onText: (t) => process.stdout.write(t),
-    onThinking: (t) => process.stdout.write(chalk.dim(t)),
+    onText: (t, complete) => process.stdout.write(t + (complete ? "\n" : "")),
+    onThinking: (t, complete) => process.stdout.write(chalk.dim(t + (complete ? "\n" : ""))),
     reasoningEffort: "medium",
     toolChoice: "auto"
 };

packages/ai/test/examples/openai-responses.ts

@@ -32,8 +32,8 @@ const context: Context = {
 }
 
 const options: OpenAIResponsesLLMOptions = {
-    onText: (t) => process.stdout.write(t),
-    onThinking: (t) => process.stdout.write(chalk.dim(t)),
+    onText: (t, complete) => process.stdout.write(t + (complete ? "\n" : "")),
+    onThinking: (t, complete) => process.stdout.write(chalk.dim(t + (complete ? "\n" : ""))),
     reasoningEffort: "low",
     reasoningSummary: "auto"
 };