[yanmxa]

Change

• Add Runtime.InterruptCurrentTurn() <-chan struct{}: request cancellation immediately and observe execution completion.
• Repeated cancellation during unwind returns the same signal.
• Keep the existing TUI InterruptTurn() 250 ms bounded wait.

Completion covers execution only, not event handling or persistence. Idle/input semantics remain unchanged; no ACP endpoint is added.

Validation

• Full Go tests, relevant race suites and lint passed; tmux startup/input/exit smoke passed.
• Production OpenAI Chat driver + local HTTP server: cancel before headers and during streaming aborts the request; completion waits for transport cleanup, with no retry or successor request observed.
• Slow-tool unwind is covered automatically. Remote provider computation/billing and other provider protocols are not verified by this HTTP test.

[san-ci]

[REVIEW NOTIFIER] This PR requires a review from @hchenxa.

Once this PR has been reviewed and has the lgtm label, it will be automatically squash-merged.

Suggested reviewer

Based on OWNERS, @hchenxa is suggested to review this PR.

How to get your PR reviewed:

• Address reviewer feedback and push changes
• Reviewer adds lgtm label when satisfied
• An approving review + lgtm will trigger auto-merge

[san-ci]

[REVIEW NOTIFIER] This PR requires a review from @hchenxa.

Once this PR has been reviewed and has the lgtm label, it will be automatically squash-merged.

Suggested reviewer

Based on OWNERS, @hchenxa is suggested to review this PR.

How to get your PR reviewed:

• Address reviewer feedback and push changes
• Reviewer adds lgtm label when satisfied
• An approving review + lgtm will trigger auto-merge

[zhujian7]

/lgtm