fix: call listener manager on_task_instance_failed() after ti state is changed

[kacpermuda]

Now we are calling get_listener_manager().hook.on_task_instance_failed with ti that still has the RUNNING state assigned. By using only ti.state, there is no way of knowing if it failed or not, it looks the same as the one being passed to get_listener_manager().hook.on_task_instance_start.

We should probably move the listener call to after, we assign the FAILED or UP_FOR_RETRY state, so that we can use that information. I've tested this change and it looks like there is no influence on any other part of code, but maybe I'm missing something.

---

^ Add meaningful description above
Read the Pull Request Guidelines for more information.
In case of fundamental code changes, an Airflow Improvement Proposal (AIP) is needed.
In case of a new dependency, check compliance with the ASF 3rd Party License Policy.
In case of backwards incompatible changes please leave a note in a newsfragment file, named {pr_number}.significant.rst or {issue_number}.significant.rst, in newsfragments.

[potiuk]

Makes sense. But would love other look :)