The newly-released
[sucker_punch](https://github.com/brandonhilkert/sucker_punch) 3.0.0
gem includes a few changes to support Ruby 3.0 keyword syntax
(see brandonhilkert/sucker_punch#235), but it
made our current tests fail in JRuby.

Relevant snippet is (from
[this failed run](https://app.circleci.com/pipelines/github/DataDog/dd-trace-rb/2764/workflows/596c0f11-c7dd-4e89-a36a-78397cc724c6/jobs/112458/parallel-runs/21)):

```
F.E, [2021-02-17T08:12:15.473545 #1277] ERROR -- : Sucker Punch job error for class: '#<Class:0x17abef0f>' args: [:fail, {}]
ArgumentError wrong number of arguments (given 2, expected 0..1)
/app/spec/ddtrace/contrib/sucker_punch/patcher_spec.rb:62:in `perform'
/usr/local/bundle/jruby/2.5.0/gems/sucker_punch-3.0.0/lib/sucker_punch/job.rb:62:in `__run_perform'
/app/lib/ddtrace/contrib/sucker_punch/instrumentation.rb:32:in `block in __run_perform'
/app/lib/ddtrace/contrib/sucker_punch/instrumentation.rb:77:in `block in __with_instrumentation'
/app/lib/ddtrace/tracer.rb:284:in `trace'
/app/lib/ddtrace/contrib/sucker_punch/instrumentation.rb:73:in `__with_instrumentation'
/app/lib/ddtrace/contrib/sucker_punch/instrumentation.rb:21:in `__run_perform'
/usr/local/bundle/jruby/2.5.0/gems/sucker_punch-3.0.0/lib/sucker_punch/job.rb:38:in `block in perform_async'
/usr/local/bundle/jruby/2.5.0/gems/concurrent-ruby-1.1.8/lib/concurrent-ruby/concurrent/executor/java_executor_service.rb:79:in `run'

Failures:

  1) sucker_punch instrumentation failed job should instrument a failed job
     Failure/Error: expect(job_span).to have_error_type('ZeroDivisionError')
       expected Span with error type "ArgumentError" to have error type "ZeroDivisionError"
     # ./spec/ddtrace/contrib/sucker_punch/patcher_spec.rb:100:in `block in <main>'
     # ./spec/ddtrace/contrib/sucker_punch/patcher_spec.rb:23:in `block in <main>'
     # /usr/local/bundle/jruby/2.5.0/gems/webmock-3.11.2/lib/webmock/rspec.rb:37:in `block in <main>'
```

An unexpected `ArgumentError` is occuring in JRuby, making our testcase
fail (because it was actually trying to cause an exception to see if it
was properly traced).

This issue is caused by a bug in sucker_punch's delegation which then
triggered a JRuby/TruffleRuby bug. I've reported the bug to the
appropriate upstreams (see #1369), and this workaround makes our CI
happy again so that we don't need to wait for upstream to react.

Fixes #1369