Lfmcmc rand engine

[gvegayon]

The LFMCMC function no longer needs an epiworld_model object to be instantiated. A rand engine is automatically created when NULL is passed instead. The data type was changed to double (from int) to give more flexibility. I added a new test that uses LFMCMC to estimate the mean and standard deviation of a normal vector. Finally, I modified the vignette to have a better kernel function (does a better job now!).

Function Enhancements:

• Modified the LFMCMC function to accept a NULL model, creating a new random-number generator engine if no model is provided. (R/LFMCMC.R, man/LFMCMC.Rd) [1] [2]
• Updated the run_lfmcmc function to include parameter type conversions and ensure compatibility with the new LFMCMC implementation. (R/LFMCMC.R)

C++ Code Updates:

• Changed the data type in the C++ code from int to double for better precision and compatibility with the new LFMCMC features. (src/cpp11.cpp, src/lfmcmc.cpp) [1] [2]
• Added logic to the LFMCMC_cpp function to handle the creation of a new random-number generator engine when the model is NULL. (src/lfmcmc.cpp)

Testing and Documentation:

• Adds new test to file test-lfmcmc.R to validate the functionality of the LFMCMC algorithm with different models and ensure accuracy. (inst/tinytest/test-lfmcmc.R)
• Updated the vignette likelihood-free-mcmc.Rmd to reflect changes in the LFMCMC setup and usage examples. (vignettes/likelihood-free-mcmc.Rmd) [1] [2]
• Adds Andrew Pulsipher as co-author of package

These changes collectively enhance the epiworldR package by providing more robust and flexible tools for model parameter estimation and simulation.

[apulsipher]

I love the purpose of this PR to allow instantiation without a model. I've two questions/changes to address here.

Additionally, the PR is failing most of the checks. Changing the data type from int to double causes unit tests to fail and causes the examples at the top of the .R file to throw errors. Please fix those before merging!

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 34.82%. Comparing base (21d89c7) to head (c96435d).
Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #51      +/-   ##
==========================================
+ Coverage   34.38%   34.82%   +0.44%     
==========================================
  Files          36       36              
  Lines        1908     1921      +13     
==========================================
+ Hits          656      669      +13     
  Misses       1252     1252              

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

---

🚨 Try these New Features:

• Flaky Tests Detection - Detect and resolve failed and flaky tests

[apulsipher]

I resolved the issues we discussed. @gvegayon Can you check that I've added myself as author correctly before merging?

[gvegayon]

I resolved the issues we discussed. @gvegayon Can you check that I've added myself as author correctly before merging?

LGTM!