Mark the hardcoded hybrid executors as deprecate and multi-exec as stable

[o-nikolas]

In preparation of deprecating support in Airflow core for the old hardcoded hybrid executors in V3, mark them as deprecated and emit warnings when they are loaded. Also mark multi exec as stable to give users somewhere to migrate towards.

To broadcast the message of their deprecation this PR:
- Adds a deprecation warning when these old hybrid executors are loaded
- Documentation has been updated to note that they are deprecated and provide steps for migration.
- A news fragment was added as well.

The warning message:

Rendered docs:

---

^ 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.

[o-nikolas]

CC: @jedcunningham @jscheffl @amoghrajesh

PR to mark the static executors as deprecated in Airflow core and mark multi-exec as stable. Re: https://lists.apache.org/thread/qmdf5dx71lzcvf4fd7qyc9gq0z0j2jqr

[jscheffl]

Looks good in my view

[o-nikolas]

Getting the tests to ignore the deprecation warning at the right time has proven very annoying/tricky.

The last stubborn one is related to one instances of the warning not being thrown, but it seems that some underlying failure is actually happening, but I'm not sure why it's happening on this branch that should be otherwise stable. And I also cannot produce it locally. From the test failure logs:

                with pytest.warns(RemovedInAirflow3Warning):
>                   file_handler.read(ti)
E                   Failed: DID NOT WARN. No warnings of type (<class 'airflow.exceptions.RemovedInAirflow3Warning'>,) were emitted.
E                    Emitted warnings: [].

tests/utils/test_log_handlers.py:289: Failed
---------------------------- Captured stdout setup -----------------------------
[2025-02-21T01:46:21.370+0000] {dagbag.py:588} INFO - Filling up the DagBag from /dev/null
----------------------------- Captured stdout call -----------------------------
[2025-02-21T01:46:21.386+0000] {dag.py:3239} INFO - Sync 1 DAGs
[2025-02-21T01:46:21.391+0000] {dag.py:3262} INFO - Creating ORM DAG for dag_for_testing_multiple_executors
[2025-02-21T01:46:21.397+0000] {dag.py:4180} INFO - Setting next_dagrun for dag_for_testing_multiple_executors to 2016-01-01T00:00:00+00:00, run_after=2016-01-02T00:00:00+00:00
[2025-02-21T01:46:21.436+0000] {file_task_handler.py:623} ERROR - Could not read served logs
Traceback (most recent call last):
  File "/opt/airflow/airflow/utils/log/file_task_handler.py", line 599, in _read_from_logs_server
    log_type = LogType.TRIGGER if ti.triggerer_job else LogType.WORKER
  File "/usr/local/lib/python3.9/site-packages/sqlalchemy/ext/associationproxy.py", line 193, in __get__
    return inst.get(obj)
  File "/usr/local/lib/python3.9/site-packages/sqlalchemy/ext/associationproxy.py", line 575, in get
    target = getattr(obj, self.target_collection)
  File "/usr/local/lib/python3.9/site-packages/sqlalchemy/orm/attributes.py", line 487, in __get__
    return self.impl.get(state, dict_)
  File "/usr/local/lib/python3.9/site-packages/sqlalchemy/orm/attributes.py", line 959, in get
    value = self._fire_loader_callables(state, key, passive)
  File "/usr/local/lib/python3.9/site-packages/sqlalchemy/orm/attributes.py", line 995, in _fire_loader_callables
    return self.callable_(state, passive)
  File "/usr/local/lib/python3.9/site-packages/sqlalchemy/orm/strategies.py", line 863, in _load_for_state
    raise orm_exc.DetachedInstanceError(
sqlalchemy.orm.exc.DetachedInstanceError: Parent instance <TaskInstance at 0x7fe810c4d6d0> is not bound to a Session; lazy load operation of attribute 'trigger' cannot proceed (Background on this error at: https://sqlalche.me/e/14/bhk3)

[amoghrajesh]

LGTM

[eladkal]

Mark the hardcoded hybrid executors

We do need to actually deprecate them in the providers code don't we?

[eladkal]

We need here also announcement that hybrid executor feature is now stable.

[o-nikolas]

Good call. I've updated the newsfragment to make that more clear!

[o-nikolas]

Mark the hardcoded hybrid executors

We do need to actually deprecate them in the providers code don't we?

That one is less of a AF3 time crunch as it can be done at any time, and I actually also think it's fine to allow the provider to improve and support those executors into the near future. Since many people will continue to use them on AF2 for a long time to come. But yes, eventually they will need to be deprecated there as well as the user base shrinks.

[jscheffl]

FYI I assume the failed CI run goes away on re-base... or as it is on v2-10-test you might need to cherry-pick #46980

Note I just realize that the PR target is v2-10-stable, I think this should rather be directed to v2-10-test!

[o-nikolas]

FYI I assume the failed CI run goes away on re-base... or as it is on v2-10-test you might need to cherry-pick #46980

Yupp, we should cherry-pick your change into v2-10-test

Note I just realize that the PR target is v2-10-stable, I think this should rather be directed to v2-10-test!

Also, yup, I accidentally targeted the wrong branch 🤦 I just chatted with @jedcunningham about it today, he's going to handle the mismatch during the next release.

[o-nikolas]

@jscheffl I added the backport-to-v2-10-test label to your PR to have it cherry picked back