While we would now build CYaml statically, we would fail to indicate to
clang the library was meant to be used statically and not dynamically.
This cleans up some linker warnings when building SPM.

Bumps [activesupport](https://github.com/rails/rails) from 6.1.7.3 to 6.1.7.6.
- [Release notes](https://github.com/rails/rails/releases)
- [Changelog](https://github.com/rails/rails/blob/v7.0.7.2/activesupport/CHANGELOG.md)
- [Commits](rails/rails@v6.1.7.3...v6.1.7.6)

---
updated-dependencies:
- dependency-name: activesupport
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

Add TopLevelDecoderTests.testDecodeOptionalTypes() to reproduce #301

Update ConstructorTests.testNull() to include assertions about handling of null specifiers that are wrapped in quotes

Update NSNull.construct(from:) to only return NSNull if the style is .plain since non-plain style scalars don't represent null (fixes bug)

Update assertion in NodeTests to explicitly require .plain style when making assertion

Correct SwiftLint violations

Add changelog entry

* [CI] Add Swift 5.9 CI jobs

* fixup! [CI] Add Swift 5.9 CI jobs

* xcodebuild_Ventura

* Xcode_Ventura

Technically, the two bug fixes could also be breaking changes in the
sense that if consumers are relying on the previously buggy behavior,
they'll now break with these fixes, but I think it's low-risk enough to
avoid bumping the major version number.