Bumps [html-proofer](https://github.com/gjtorikian/html-proofer) from 3.17.3 to 3.17.4.
- [Release notes](https://github.com/gjtorikian/html-proofer/releases)
- [Commits](https://github.com/gjtorikian/html-proofer/compare/v3.17.3...v3.17.4

)

Signed-off-by: dependabot[bot] <support@github.com>

dev-cmd/dispatch-build-bottle: set ARM labels.

Set the label properly for ARM.

Tested with inputs `11-arm` and `arm64_big_sur`.

Fixes https://github.com/Homebrew/brew/issues/9434

Fix `:arm64_big_sur` bottle ordering.

brew.sh: don't allow system Ruby on Catalina.

livecheck: add GithubLatest strategy

This wasn't being parsed correctly so was being put below the Linux
bottle which is less readable, less intuitive and will cause a bunch
of merge conflicts.

It's (far) too broken for our purposes.

Fixes https://github.com/Homebrew/brew/issues/9410

workflows/tests: consolidate macOS tests.

git_extensions: refactor and delete redundant functions

Merge pull request #9431 from RandomDSdevel/restore-semantically-significant-indentation-lost-accidentally-in-PR-9398

[Library/Homebrew/cli/named_args.rb] `to_formulae_to_casks()`:  Reasssociate its `rescue` Block with the Correct Enclosing Scope

`rescue` block with the right enclosing scope.

     PR #9398 accidentally removed some semantically significant indentation in
this function, thus inadvertently kicking the `rescue` block inside out to func-
tion scope, when it moved its code further up in the file.  This may well not
confuse a Ruby interpreter, but it could trip human readers up.  Compensate and
correct the overlooked offending oversight and potential refactoring mistake by
restoring only that indentation, leaving the function in its new location.

Refs #9398.

brew: re-exec under native arch if requested

Ensure default cask tap is installed on `CaskUnavailableError`.

Test for `brew ruby -e 'puts "testball".f.path'`

Remove style exceptions for ProvidedByMacos and UsesFromMacos

Co-authored-by: Mike McQuaid <mike@mikemcquaid.com>

Add CPU family for M1, remove Intel Arrandale

According to reports, XNU’s CPU family identifier previously used for
the Westmere family (which Homebrew mislabels as the Arrandale family)
is now being reused for one of Apple’s M1 SoC models.

Resolve the conflict by splitting the list of CPU families and adding a
microarchitecture check.

Sources for the values:

- https://github.com/Homebrew/brew/issues/7857#issuecomment-728739049
- https://en.wikipedia.org/wiki/Comparison_of_ARMv8-A_cores
- Direct messages

build(deps): bump rspec-sorbet from 1.7.0 to 1.8.0 in /Library/Homebrew

We have a limited number of macOS workers (for the Homebrew
organisation) and queuing these seems to be slowing things down overall
compared to when we had one big job.

Linux throughput is still significantly quicker so it makes sense to
split up those jobs for providing quick feedback on breakages that are
cross-platform (which ends up being most but definitely not all of
them).

add flags to deps --installed

utils/pypi: return nil for non-pypi-packages from url