Every job on Travis is unstable now.

Those jobs are way too unstable and provide very little benefit.

e.g.
https://app.travis-ci.com/github/ruby/ruby/jobs/624611955
https://app.travis-ci.com/github/ruby/ruby/jobs/625551620

I <jaruga@ruby-lang.org> haven't received the notifications since August 2022.
So, try another way to be notified written on the document below.

https://docs.travis-ci.com/user/notifications/#configuring-email-notifications

"Not document only" should be honored always.  Also simplified repo
and branch conditions

As Travis CI IBM z pipeine is operational again, I will add s390x again to check
the case. Though Travis s390x infra team is still investigating the root cause.
https://www.traviscistatus.com/

Please revert this commit or comment on the
<https://bugs.ruby-lang.org/issues/20013>, when you see the s390x builds are not
starting again.

Sorry for inconvenience.

The builds are not starting.

You can check the Travis status on the ticket,
<https://bugs.ruby-lang.org/issues/20013>.

https://app.travis-ci.com/github/ruby/ruby/builds/267410310

Checking the past 10 builds in Travis CI arm64, it looks stable. So, disable
`allow_failures`.

Also sorted the order of the jobs for the priority. The arm64 is the first. And
I would like to run arm32 using the arm64 pipeline a bit later after the
arm64 using the same arm64 pipeline. So, the arm32 is the last.

In the past, I saw that the 2nd arm64 pipeline was unstable. Now I want to try
it to check if it is stable enough to run again.

As far as I know, right now the only native arm64 cases are MacOS Arm
`macos-arm-oss` on GitHub Actions. And I want to add Linux arm64 case on
pull-request in CI.

Add a RubyGems issue ticket link. The issue happend in Travis arm32 at once.