Improve Benchmark consistency and visibility

I've been involved in a cross-language initiative to improve our community's performance evaluation. Here are the things that I'm working on relating to this repo:

• add an org level shared runner that uses bare-metal to improve test consistency
  • https://github.com/open-telemetry/community/issues/1662
• add a workflow to run JMH benchmarks via github actions
  • https://github.com/open-telemetry/opentelemetry-java/pull/5842
• request access to shared runner for added workflow on main
  • https://github.com/open-telemetry/community/issues/1699
• publish results to a github pages branch
• publish results to otel website
• define cross language benchmark scenarios in spec
  • https://github.com/open-telemetry/opentelemetry-specification/pull/3696
• implement/update benchmark in java repo to match spec
• ensure benchmark execution is constrained and short to prevent shared resource abuse
  • https://github.com/open-telemetry/opentelemetry-java/pull/5842

Start by reviewing the unchecked items in the issue and the linked cross-language initiative and specification pull request. The completed workflow and benchmark work are linked from the checklist; completion would require resolving the remaining website publication, scenario definition, and Java benchmark alignment work.