Fix fd leak when restarting external validator

This MR fixes an fd leak that occurs when the external validator is automatically restarted by the node. The communication socket of the former (crashed) validator was left open and discarded. We now close it properly.

Manually testing the MR

Run a sandbox node. Activate alpha.

• Use lsof -p NODE_PID | wc -l to count the number of fds opened by the node. lsof -a -p NODE_PID can also be useful to see that multiple "validator-socket" as kept open.
• kill the octez-validator using kill -9 VALIDATOR_PID (the pid is given in the node's logs)
• octez-client bake for --minimal-timestamp
• notice a fd bump of 2: - 1 is the validator fd that is not close; - 1 is another fd leak spotted by this investigation that will be tackled later.

With the patch, the number of fds is increased by 1 only (because of the other fd leak.)

Checklist

• Document the interface of any function added or modified (see the coding guidelines)
• Document any change to the user interface, including configuration parameters (see node configuration)
• Provide automatic testing (see the testing guide).
• For new features and bug fixes, add an item in the appropriate changelog (docs/protocols/alpha.rst for the protocol and the environment, CHANGES.rst at the root of the repository for everything else).
• Select suitable reviewers using the Reviewers field below.
• Select as Assignee the next person who should take action on that MR