Parametrize struct definitions of random matrix ensemble types

[apkille]

See #91.

[codecov]

Codecov Report

Attention: Patch coverage is 90.00000% with 3 lines in your changes missing coverage. Please review.

Project coverage is 42.65%. Comparing base (37f177b) to head (6f55049).

Files with missing lines	Patch %	Lines
src/GaussianEnsembles.jl	88.00%	3 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master      #96   +/-   ##
=======================================
  Coverage   42.65%   42.65%           
=======================================
  Files          12       12           
  Lines         830      830           
=======================================
  Hits          354      354           
  Misses        476      476           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[apkille]

@dlfivefifty ping for review

[dlfivefifty]

I’m not sure this change is a good idea since GUE, GOE, and GSE are very different and should be different types.

for general β this makes sense. Though my personal preference would be to use Unicode

[apkille]

That's fair, I'll revert this. I guess with that in mind, would it be good to parametrize other ensemble types this way, e.g. Ginibre{4}?

[apkille]

@dlfivefifty

[dlfivefifty]

I think so. Especially if the code consists of a lot of special cases if β == 1, elseif β ==2, elseif β == 4 end then that hints that they actually should be treated as different types via parameterisation

[apkille]

OK sounds good.