Port PSO algorithm from legacy Funz plugin to new fz S3 class format

[Copilot]

Migrates the Particle Swarm Optimization algorithm from the old Funz plugin architecture to the new fz framework's S3 class pattern.

Implementation

• S3 class structure: Constructor returns classed object with options list and mutable state environment
• Method dispatch: getInitialDesign/getNextDesign/displayResults → get_initial_design.PSO/get_next_design.PSO/get_analysis.PSO/get_analysis_tmp.PSO
• Data format: Matrix-based I/O → list of named lists (e.g., list(list(x1=0.5, x2=0.3), ...))
• Analysis return: HTML string → structured dict with text, data, html fields
• Termination signal: NULL → empty list list()
• Variable scoping: Explicit capture for future block execution (id, d, seed, maxit)

Example

# Old format
pso <- PSO(list(maxit='30', seed='123'))
X <- getInitialDesign(pso, input, output)  # Returns matrix

# New format  
pso <- PSO(maxit=30, seed=123)
X <- get_initial_design(pso, input_vars, output_vars)  # Returns list of named lists
results <- get_analysis(pso, X, Y)  # Returns list(text=..., data=..., html=...)

Additional files

• tests/test_basic.R: S3 structure validation
• examples/branin_example.R: Complete optimization workflow
• PORTING_GUIDE.md: Migration reference for other algorithms
• .github/workflows/test.yml: CI with explicit permissions

Original prompt

Create a repository porting old PSO plugin (https://github.com/Funz/algorithm-PSO) to new fz. See https://github.com/Funz/fz/blob/implement-algorithms/examples/algorithms/montecarlo_uniform.R to get details of new fz "algorithm" in R (mostly files in .fz/algorithms/...)

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.