Use touchstone

[VisruthSK]

Use touchstone to evaluate PRs' impact on performance. Starting with a monolithic approach to testing, calling just loo().

Closes #348.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 92.80%. Comparing base (60f012d) to head (c73be4d).
⚠️ Report is 3 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #352   +/-   ##
=======================================
  Coverage   92.80%   92.80%           
=======================================
  Files          31       31           
  Lines        3004     3004           
=======================================
  Hits         2788     2788           
  Misses        216      216           

☔ 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.

[VisruthSK]

Should lower reps to maybe 5, perhapsalso use smaller data? Current ones are fully LLM generated so maybe more careful creation could cut down on execution time while still being informative tests.

@jgabry any thoughts? The main file is touchstone/script.R The code rn is ugly, I'll fix it tomorrow probably, but any thoughts on the actual tests? The run is taking 1hr, is that way too slow?

[jgabry]

If we switch to the example that @avehtari suggested, is that a bigger log lik matrix/array than what you have here or smaller?