[Bug]: health check continues to work after the instance has been stopped

[TheBits]

dstack version

0.16.2

Python version

3.11.8

Host OS

Arch Linux

Host Arch

x86_64

What happened?

We need to stop run the health_check on the terminating instances.

When the instance terminates, the health check continues to function. Sometimes the health check sets an incorrect status to the table.

termination_reason	health_status
Idle timeout	Can't request shim: ('Connection aborted.', ConnectionResetError(104, 'Connection reset by peer'))

CLI logs

No response

Server logs

No response

Runner logs

No response

Additional Information

No response