Change CommandInfoCache to implement IDisposable and clean up the runspace pool

[JamesWTruher]

PR Summary

The Helper is recreated with every invocation of Invoke-ScriptAnalyzer which essentially results in a runspace which was not cleaned up for every invocation. I saw 100s of runspaces (ia get-runspace) in my session. I also gave the CommandInfoCache it's own runspace pool and implemented IDisposable in it. Lastly, I've added a test to catch this in the future.

PR Checklist

• PR has a meaningful title
  • Use the present tense and imperative mood when describing your changes
• Summarized changes
• Change is not breaking
• Make sure all .cs, .ps1 and .psm1 files have the correct copyright header
• Make sure you've added a new test if existing tests do not effectively test the code changed and/or updated documentation
• This PR is ready to merge and is not Work in Progress.
  • If the PR is work in progress, please add the prefix WIP: to the beginning of the title and remove the prefix when the PR is ready.

[bergmeister]

The new test failed on WMF4 because the Get-Runspace Cmdlet is not available in that PS version. We could either exclude this PS version for that test (given the fix is ps version agnostic I think that would be OK) or maybe there is a lower level method that one can call as an alternative?

This fix shows up implicitly in the build times, which are quite a bit faster when compared to the ones of master (1 minute), and the WMF4 one even 6 minutes. I even remember being realizing that the WMF4 builds started to take quite a bit longer at some point but I just thought that AppVeyor had reduced the VM sizes due it being a legacy image.

I added a few suggestions and have just one question around the maximum number of runspaces to assert against but mostly this PR looks great.

[bergmeister]

Looks ok, can be merged if the build is green (except for the Ubuntu build)