Re-organise tests.

[yebai]

This PR makes it easy to run each test file individually locally:

shell> pwd
/Users/hg344/projects/Turing.jl

(test) pkg> activate test
  Activating project at `~/projects/Turing.jl/test`

julia> using Turing; include(pkgdir(Turing)*"/test/test_utils/test_utils.jl")

julia> include("test/modes/OptimInterface.jl")

This PR doesn't contain any functionality changes.

[torfjelde]

Do we feel this is more readable than what we did before?

As in, why don't we just keep the separate files but include all of them in a single file?

[yebai]

I suspect that anyone ever reads them.

I prefer to keep these code in fewer files since it often contains a collection of tiny utility sets.

[torfjelde]

I'm uncertain if this is an issue of Turing.jl or rather an issue of "standard Julia workflow".

What I usually do is that I take this "header" for the runtests.jl, paste it into a setup.jl file (similar to what you've done, but using include instead), and then use this is a setup-file for my testruns. I then use a custom script which re-runs the files I've specified whenever changes are made to either the package or the tests themselves (script: https://gist.github.com/torfjelde/62c1281d5fc486d3a404e5de6cf285d4). You might find this useful:)

[yebai]

@torfjelde Your workflow is much better! However, it is helpful to introduce a simple way to achieve a similar objective. Hence this PR.

[torfjelde]

However, it is helpful to introduce a simple way to achieve a similar objective. Hence this PR.

Fair! But do we need to also remove all the files in the process? As in, why not just use include but add a single file that contains all these?

[coveralls]

Pull Request Test Coverage Report for Build 5719263514

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 0.0%

---

Totals
Change from base Build 5718410481:	0.0%
Covered Lines:	0
Relevant Lines:	1463

---

💛 - Coveralls

[codecov]

Codecov Report

Patch and project coverage have no change.

Comparison is base (e22b77c) 0.00% compared to head (570fee5) 0.00%.
Report is 1 commits behind head on master.

Additional details and impacted files

@@          Coverage Diff           @@
##           master   #2058   +/-   ##
======================================
  Coverage    0.00%   0.00%           
======================================
  Files          21      21           
  Lines        1447    1447           
======================================
  Misses       1447    1447           

Files Changed	Coverage Δ
src/Turing.jl	0.00% <ø> (ø)
src/mcmc/Inference.jl	0.00% <ø> (ø)
src/mcmc/abstractmcmc.jl	0.00% <ø> (ø)
src/mcmc/emcee.jl	0.00% <ø> (ø)
src/mcmc/ess.jl	0.00% <ø> (ø)
src/mcmc/gibbs.jl	0.00% <ø> (ø)
src/mcmc/gibbs_conditional.jl	0.00% <ø> (ø)
src/mcmc/hmc.jl	0.00% <ø> (ø)
src/mcmc/is.jl	0.00% <ø> (ø)
src/mcmc/mh.jl	0.00% <ø> (ø)
... and 3 more

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

[yebai]

Closed in favour of #2179