Turns out this is important for the black formatting tool to work
correctly. The formatting won't generally change between python versions
(it only depends on `black`'s version), but the formatted code needs to
be parseable by the system python version. One script uses
```python
def foo[T](x: T) -> T:
```
syntax, which is only supported in Python 3.12 and later.
Running `pre-commit` will now require a python 3.12 installation (which
is already what we mandate for internal developer environment setup).
The error in case of absence of such a version is pretty clear though.
Seems like `windows-latest-xl` is not available any more. This should unblock CI, but longer term we should consider doing what other languages do (i.e. run tests from the internal repo).
* remove ql test running and upgrade/downgrade scripts checking (now
done internally)
* removed all the bazel caching stuff, that never really worked any way
* moved `misc/codegen` generic testing to a separate workflow, as it's
not swift specific any more
* reinstanted checking that the extractor can be built locally from
the `codeql` repo.
These are test cases for the GitHub Actions analysis.
Exclude them when running code scanning against this repo,
to avoid noisy alerts.
Test workflow files in this directory are safe from
execution, because Actions only executes workflows
that live directly in the .github/workflows
top-level directory.
`action.yml` files in this directory can in theory
be executed as a step in a workflow; for now exclude them.
`cargo fmt --all` should be avoided in CI as that will include `rust/ast-generator`
which has sources provided by bazel (`bazel run //rust/ast-generator:inject_sources`
can provide those sources in-tree).
Now the formatting checks are limited to the sources that trigger the jobs, and a
check is added to `rust/ast-generator`.
