Python 3.8 [reaches EOL in
October](https://devguide.python.org/versions/). Meanwhile, python 3.11
and 3.12 have been released since we last updated our supported python
versions. This PR adds support for 3.11 and 3.12, and changes the
default bazel targets to use 3.9 instead of 3.8. Some things of note:

- The bazel dependency for torchvision in python3.10 is no longer
resolving (and doesn't resolve for 3.11 or 3.12 either). For now, this
PR handles that by making all bazel targets depending on torch vision
only get built/tested for 3.8 and 3.9 . Note that this does NOT mean the
examples using torch vision will not be runnable in higher python
versions outside the context of bazel.

- recent `rules_python` introduce a [complex bazel workspace dependency
issue](bazel-contrib/rules_python#1543) so we can
only have bazel targets for building and testing with interpreters up to
python3.11. Note again that this is a limitation of what we can do *in
bazel* and not a limitation of what interpreters the resulting wheel can
be run with.

Note that this does not change what version of python the server or
default worker images run with (python3.9 for both). So this is
essentially an sdk and local-server only change.

Testing
--------

- Ran all bazel tests
- Used bazel CLI to start local server and run the MNIST example against
it
- built a wheel and ran the following tests on fresh virtual envs with
python 3.9/3.10/3.11/3.12: (a) install from wheel (b) stop and start
local server (c) run mnist example (d) view it in browser

---------

Co-authored-by: Josh Bauer <josh@sematic.dev>

SQLAlchemy 2.0 came out in January of 2023 (>1.5 years ago). It
introduced a variety of breaking changes. Sematic uses SQLAlchemy
heavily, and in complex ways (ORM type introspection for dynamic
serialization/deserialization, ORM type mixins, complex and dynamic
query builds). This meant we had to wait to upgrade until sufficient
time could be dedicated to migrating and testing it. We don't want to
wait too much longer, it's only a matter of time before 1.x releases
stop getting published and SQLAlchemy's development has largely shifted
to >=2.0. So this PR bites the bullet and does the upgrade.

This PR also upgrades the server image python version to 3.12. While not
technically related to the SQLAlchemy changes, it follows closely on the
heels of #1126 , and since that did not change anything running on the
server (but this does) it seemed like an OK place to add so the 3.12
server and SQLAlchemy 2.0 changes could be tested together.

Finally, there was actually a bug for custom metrics usage that was not
caught in unit tests. This PR fixes that bug, confirmed in remote
testing.

Testing
--------

In addition to all unit tests passing, I also tested running a local
server with bazel and running the MNIST example in it. I confirmed
built-in metrics worked and the pipeline passed with the expected
resulting UI.

I then built a wheel with these changes and published a release
candidate Docker image for the standard and EE images.

In a remote environment, I:

- Confirmed the built-in system pipeline metrics (ex: success rate and
timings) worked
- ran a pipeline end-to-end
- ran a pipeline that used a Ray cluster
- ran a pipeline that used custom metrics and confirmed the UI works
- cofirmed that the cleaner is working
- tested run search extensively

---------

Co-authored-by: Josh Bauer <josh@sematic.dev>

Release SQLAlchemy version change and python 3.11 / 3.12 support

Co-authored-by: Josh Bauer <josh@sematic.dev>