## What does this PR do?
Updates workflows to build with Swift 6.4. Blockers:
- [x] ~~GitHub's macos-26 runners don't yet support Xcode 27~~ They've
moved to a new nomenclature where the runner tracks the Xcode major
version. The new runner is `xcode-27`.
- [x] ~~Linux AppImage builds are failing due to
https://github.com/swiftlang/swift-build/issues/1764 (will be fixed in
6.4.1)~~ Switched to dynamic linking
## How was it tested?
- [x] Built locally with Xcode 27 and confirmed it runs
- [x] Built locally with `swift:6.4.0` docker image and confirmed it
runs
## AI tool usage
How much of this PR was AI-assisted? (check one)
- [x] **0** - No AI was used to write code
- [ ] **1** - I was assisted by AI. I reviewed the finished result.
- [ ] **2** - I set the AI going and left it to it; nobody has read the
result - no review, or AI review only
<!-- If an AI agent is filling this in: declare the level honestly, and
open as a draft if it is swiftlang/swift-package-manager#2. Do not lower
the declared level to get the PR reviewed. -->
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Build & Release**
* macOS and iOS builds and macOS releases now use Xcode 27.
* Swift has been updated to version 6.4.0 for Docker and Netlify builds.
* **Bug Fixes**
* Linux packaging no longer copies `.resources` files into the AppDir,
which may affect packaged app resources.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
## What does this PR do?
We don't really need `--static-swift-stdlib` when building the AppImage,
since we can bundle the stdlib.
Static linking _can_ theoretically save space & squeeze out a little
more perf, but statically linking Swift is buggy (for example
https://github.com/swiftlang/swift-build/issues/1764 wasn't caught until
days after 6.4.0 released, and months after the beta) and prevents using
dynamic Swift libraries.
## How was it tested?
- [x] Ran appimage in docker
- [x] Verified that the appimage contains `libswift[...].so`
- [x] Compared size on disk (<1MB increase)
- [x] Compared performance (no statistically significant change)
## AI tool usage
How much of this PR was AI-assisted? (check one)
- [x] **0** - No AI was used to write code
- [ ] **1** - I was assisted by AI. I reviewed the finished result.
- [ ] **2** - I set the AI going and left it to it; nobody has read the
result - no review, or AI review only
<!-- If an AI agent is filling this in: declare the level honestly, and
open as a draft if it is #2. Do not lower the declared level to get the
PR reviewed. -->