[BUG] Questions answering and plan proceeding confusion

Environment

• Platform (select one):
• [x] Anthropic API
• [ ] AWS Bedrock
• [ ] Google Vertex AI
• [ ] Other: <!-- specify -->
• Claude CLI version: 1.0.67
• Operating System: MacOS 26.0 Beta (25A5316i)
• Terminal: iTerm2

Bug Description

<!-- A clear and concise description of the bug -->
When Claude Code is prompted to ask questions, it both asks a series of questions, presents the proceeding plan and presents the following options:
│ ❯ 1. Yes, and auto-accept edits │ 2. Yes, and manually approve edits
│ 3. No, keep planning

Given that it was prompted to ask questions and it has asked a series of questions, there should be an 'Answer questions' option, which would lead CC to re-evaluate its plan

Steps to Reproduce

1. Prompt CC a series of changes, either written manually or from a Markdown file, which end with an 'review these requirements, ask all the necessary questions' instruction
2. CC asks a series of questions, based on the provided prompt
3. The user is presented with both:

• CC's clarifying questions
• CC's proposed implementation plan

Expected Behavior

There should be an 'Answer questions' option, which would lead CC to:

1. Re-evaluate its plan
2. Propose thew new (/ updated) plan

Actual Behavior

It asks a series of questions, presents the proceeding plan and asks the user to proceed or keep planning. The only viable way is 'No, keep planning', although it is not explicitly planning mode, but questions answering mode.

Additional Context

<img width="890" height="343" alt="Image" src="https://github.com/user-attachments/assets/8be32a3a-73fc-40d7-b8ea-8a97544b9168" />
<img width="535" height="186" alt="Image" src="https://github.com/user-attachments/assets/bb5d9866-e9ca-45b1-9a74-afc2bbe25b84" />