This is the last big change in "Rewrite Breeze to Python".

Fixes: #23538

Co-authored-by: Anthony Panat <anthonypanat@Anthonys-MacBook-Pro-2.local>
Co-authored-by: Anthony Panat <anthonypanat@anthonys-mbp-2.mynetworksettings.com>

Co-authored-by: ok9897 <ok9897@krafton.com>

* Added information about Amazon Elastic MapReduce Connection

* Sagemaker System Test fixes

* Add a missing file.

* Fix example_cloud_build.py

Co-authored-by: yanmulin <lyuweny@gmail.com>

Signed-off-by: Ross Turk <ross@rossturk.com>

Signed-off-by: Ross Turk <ross@rossturk.com>

As part of 2.3.3 we added redaction to output from the tasks test
command, but unfortunately that broke using a debugger with this error:

```
  File "/usr/lib/python3.10/pdb.py", line 262, in user_line
    self.interaction(frame, None)
  File "/home/ash/.virtualenvs/airflow/lib/python3.10/site-packages/pdb.py", line 231, in interaction
    self._cmdloop()
  File "/usr/lib/python3.10/pdb.py", line 322, in _cmdloop
    self.cmdloop()
  File "/usr/lib/python3.10/cmd.py", line 126, in cmdloop
    line = input(self.prompt)
TypeError: 'NoneType' object cannot be interpreted as an integer
```

(ipdb has a similar but different error)

The "fix" is to assign a fileno attribute to the object. `input()` needs
this to write the prompt. It feels like a "bug" that it doesn't work
without it, but as this class is only used in `tasks test` this is a
safe change

This warning was invisible before 2.4 due to a bug in our logging config
(fixed by commit 7363e35c) and AIP-45 which suddenly made this appear.

The problem was being caused by set_xcomargs_dependencies being called
once for each class in the hierarchy, and each of them doing the same
logic.

The fix is to look at the _actual_ function of `self.__init__` and
compare it to the function we're about to call so that we don't set
dependencies until we have finished the "outer" most class's
apply_defaults invocation.

Fix separator getting added to variables_prefix

azure key vault adds a - between the key prefix and the secret name when it reads it from azure key vault. It does this even when the key prefix is "" so secrets end up named -my-secret instead of my-secret.

* Ensure the log messages from operators during parsing go somewhere

While investigating #26599 and the change from AIP-45, I noticed that
these warning messages weren't new! The only thing that was new was that
we started seeing them.

This is because the logger for BaseOperator and all subclasses is
`airflow.task.operators`, and the `airflow.task` logger is not
configured (with `set_context()`) until we have a TaskInstance, so it
just dropped all messages on the floor!

This changes it so that log messages are propagated to parent loggers by
default, but when we configure a context (and thus have a file to write
to) we stop that. A similar change was made for the `airflow.processor`
(but that is unlikely to suffer the same fate)

* Give a real row count value so logs don't fail

The ArangoDB sensor test was logging a mock object, which previously was
getting dropped before emitting, but with this change now fails with
"Mock is not an integer" when attempting the  `%d` interpolation.

To avoid making the mock overly specific (`arangodb_client_for_test.db.`
`return_value.aql.execute.return_value.count.return_value`!) I have
changed the test to mock the hook entirely (which is already tested)

This makes the starting of StandardTaskRunner faster as the parsing of DAG will now be done once at task_run.
Also removed parsing of example dags when running a task

Previously the code used the dags_folder of the "current" process (e.g. the celery worker, or k8s executor worker pod) to calculate the relative fileloc based on the full fileloc stored in the serialized dag.  But if the worker dags_folder folder is different from the dags folder configured on the dag processor, then airflow can't calculate the relative path, so it will just use the full path, which in this case will be a bad path.  We can fix this by keeping track of the dags_folder from the dag processor that serialized the dag, and using this for figuring out the relative path.

* Add all dependent providers when running selective checks

Like we run tests for providers that use a changing provider, we need to
run tests for any providers we use directly.

So for example, if we change the `postgres` provider, both `common.sql`
(which `postgres` uses) and `google` (which uses `postgres`) should be
run.

Copy and paste the older code snippet into the values.yml did not work, the migrations jobs were not triggered when using ArgoCD. I had to place the code in this format in order to function.

Found during September release:

* fix process steps to be copy&pasteable
* add -m to tag generation to account for signed/annotated tags
* added twine check to the CI workflow

The sphinx-autoapi bug https://github.com/readthedocs/sphinx-autoapi/issues/352
has been resolved and 2.0.0 version has been released. Since we
are using sphinx/sphinx-autoapi only for `doc` extra, we can
safely bump the minimum version of both to latest released minor
versions.

Just don't need this line.

Previously mapped tasks each existed as their own independent-but-unlabeled points.  Makes sense to aggregate them somehow.