Start bb nREPL server at a random port

[ikappaki]

Hi,

could you please consider PR to start bb nREPL server at a random port. It fixes #3293.

The :0 format eligibility to start the server has been confirmed by @borkdude on chat.

I did not bother to include a test for such a small change, but here is the output from the bb integration test that shows port is now not 1667 but random:

[00:11.202]  [nREPL] Starting server via /opt/hostedtoolcache/Babashka/1.0.165/x64/bb nrepl-server :0
[00:11.209]  [nREPL] server started on 33711
[00:11.209]  [nREPL] Establishing direct connection to 127.0.0.1:33711 ...
[00:11.209]  [nREPL] Direct connection to 127.0.0.1:33711 established
[00:11.393]  Connected! In the absence of parentheses, chaos prevails.
[00:12.145]  [nREPL] Connection closed
[00:12.146]  nREPL server exited.
jack in
  to babashka
  to babashka (1.15s)

I've also piggybacked an update to the circleci job names to include ubuntu so names are aligned with the rest of the jobs, if you don't mind.

• The commits are consistent with our contribution guidelines
• You've added tests (if possible) to cover your change(s)
• All tests are passing (eldev test)
• All code passes the linter (eldev lint) which is based on elisp-lint and includes
  • byte-compilation, checkdoc, check-declare, packaging metadata, indentation, and trailing whitespace checks.
• You've updated the changelog (if adding/changing user-visible functionality)
• You've updated the user manual (if adding/changing user-visible functionality)

Thanks

[benjamin-asdf]

ah lol you beat me to it. But 0 also works.

[bbatsov]

Looks good to me. Thanks!

P.S. This PR reminded how useless the *-global-params variables are, as we never used them for anything. Probably we should just remove them across the board to simplify a bit the addition of new jack-in backends.