Test ansible PR 84240 wrt connection timeout option

connection_timeout.yml

- hosts: d12.mynet
  connection: ssh_noisy_get_option
  gather_facts: false
  vars:
    ansible_python_interpreter: python3
  tasks:
    - meta: reset_connection
    - ping:

ssh_noisy_get_option.py

from ansible.plugins.connection.ssh import (
    Connection as Connection_,
    DOCUMENTATION as DOCUMENTATION_,
)
from ansible.utils.display import Display


DOCUMENTATION = DOCUMENTATION_.replace(
    '    name: ssh\n',
    '    name: ssh_noisy_get_option\n',
)


display = Display()


class Connection(Connection_):
    transport = 'ssh_noisy_get_option'

    def get_option(self, option, hostvars=None):
        result = super().get_option(option, hostvars=hostvars)
        if option == 'timeout':
            display.warning(f'ssh_noisy_get_option.get_option({option!r}, ...) -> {result!r}')
        return result

tox_connection_timeout.ini

[tox]
envlist =
    py312-ansible{8,9,11,10.pr84240,11.pr84240,12.dev}
skipsdist = true

[testenv]
deps =
    ansible8: ansible~=8.0
    ansible9: ansible~=9.0
    ansible10: ansible~=10.0
    ansible11: ansible~=11.0
    ansible10.pr84240: git+https://github.com/s-hertel/ansible.git@backport/2.17/84240
    ansible11.pr84240: git+https://github.com/s-hertel/ansible.git@backport/2.18/84240
    ansible12.dev: git+https://github.com/ansible/ansible.git@01ca9b1d0ec882f5eea7fc42ef7f7dab9ea76d19
setenv =
    ANSIBLE_CONNECTION_PLUGINS={toxinidir}/lib/connection
    ANSIBLE_PIPELINING=true
    ANSIBLE_STDOUT_CALLBACK=default
commands =
    ansible-playbook -i d12.mynet, connection_timeout.yml

The overwriting behaviour I saw while working on mitogen-hq/mitogen#1201 must be specific to Mitogen