[MODEL] Subagents continue problem.

### Preflight Checklist

- [x] I have searched [existing issues](https://github.com/anthropics/claude-code/issues?q=is%3Aissue%20state%3Aopen%20label%3Amodel) for similar behavior reports
- [x] This report does NOT contain sensitive information (API keys, passwords, etc.)

### Type of Behavior Issue

Other unexpected behavior

### What You Asked Claude to Do

I have asked claude to resume a work with subagents of a conversation has been closed after I stopped the agents, and opened a new claude code in another window then /resume to that conversation.

### What Claude Actually Did

he said that he can't resume a sub agent that has been closed by the user, this is so fruastrating, as you can see in this image:

### Expected Behavior

1. Tell the subagents to continue (or you can decide more professional way)
2. take the results and continue normally.

### Files Affected

```shell

```

### Permission Mode

I don't know / Not sure

### Can You Reproduce This?

Haven't tried to reproduce

### Steps to Reproduce

1. Tell claude code to make N subagents.
2. Stop each one of them by "x" on keyboard.
3. Fully close the current claude code proccess (like in terminal window)
4. Open another window then run claude code.
5. /resume to that conversation.
6. tell him to continue and resume the subagents.

### Claude Model

Opus

### Relevant Conversation

```markdown

```

### Impact

High - Significant unwanted changes

### Claude Code Version

2.1.268

### Platform

Anthropic API

### Additional Context

_No response_

No files, tests, or entry points are identified. Start by following the listed reproduction sequence with Claude Code 2.1.268, then inspect the resume and subagent lifecycle entry points; done means a resumed conversation can continue stopped subagents and incorporate their results.

Do mô hình lập chỉ mục viết ra từ nội dung của issue.