Support a simple augmented data interface

[richfitz]

No description provided.

[edknock]

proposal is not symmetric so you need to account for this in the acceptance probability, it should be something like

proposal_ratio <- z * log(p_group_1) + (1 - z) * log(1 - p_group_1) -
        z_new * log(p_group_1) - (1 - z_new) * log(1 - p_group_1)
p_accept <- pmin(1, exp(density_new - density + proposal_ratio))

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.92%. Comparing base (53ae53f) to head (4ebd95a).

Additional details and impacted files

@@           Coverage Diff            @@
##             main     #176    +/-   ##
========================================
  Coverage   99.92%   99.92%            
========================================
  Files          70       70            
  Lines        5458     5573   +115     
========================================
+ Hits         5454     5569   +115     
  Misses          4        4            

☔ 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.

[edknock]

I have set this up so that parallel tempering or HMC do not support augmented data - they could do in future but would need a bit more of a think about how they work.