QA: Reasoning (Default Render) — Mastra

Test Steps

• Navigate to /demos/reasoning-default-render
• Ask a reasoning-prone question
• If the model emits reasoning tokens, verify the built-in CopilotChatReasoningMessage collapsible card renders

Expected Results

• No custom reasoningMessage slot is supplied by the page
• Built-in UI handles the reasoning output

Known Limitation

Same as agentic-chat-reasoning: depends on the underlying LLM emitting reasoning tokens.