BayesianNetwork 0.4

New features

• Added unit tests using testthat and shinytest2 for improved reliability.
• Added download handlers for the adjacency matrix (CSV), fitted network (BIF format), and network parameters (RDS).
• Added inline help text describing each of the 12 supported structure learning algorithms.
• Added support for multi-node evidence in inference: users can now accumulate evidence across multiple nodes before querying.
• Added Gaussian network inference using likelihood-weighting sampling (bnlearn::cpdist()), with kernel density visualization of the posterior.

Minor improvements and bug fixes

• Refactored server.R and ui.R for improved code quality.
• Updated R-CMD-check CI/CD workflow.
• Added knitr and rmarkdown vignette builder support.