MuLambda EA

[francastagna]

PR Summary

Add (μ,λ) Evolutionary Algorithm and tests

---

(μ,λ) EA Pseudocode

---

Main Changes

• EMConfig

  • Add MuLambdaEA to EMConfig.Algorithm.
  • Add muLambdaOffspringSize (λ) for (μ,λ) EA.

• Main wiring

  • Wire MuLambdaEvolutionaryAlgorithm for all problem types (GraphQL, RPC, Web, REST).

• New algorithm

  • MuLambdaEvolutionaryAlgorithm.kt
    • For each generation: generate exactly λ offspring by mutating parents (distribute as ⌊λ/μ⌋ plus remainder), then select the best μ from offspring only.
    • Lifecycle hooks: beginGeneration/endGeneration and beginStep/endStep for each offspring.

---

Tests

• MuLambdaEvolutionaryAlgorithmTest.kt
  • Finds Optimum (testMuLambdaEAFindsOptimum).
  • Edge cases:
    • xoverProbability=0.0, fixedRateMutation=1.0 → 0 crossovers, λ mutations, µ size preserved. (testNoCrossoverWhenProbabilityZero_MuLambdaEA)
    • fixedRateMutation=0.0, xoverProbability=1.0 → 0 mutations, µ size preserved. (testNoMutationWhenProbabilityZero_MuLambdaEA)
  • One-iteration properties (testNextGenerationIsTheBestMuFromOffspringOnly):
    • Final population size equals µ.
    • Final population equals top-µ by score from offspring only.
    • With fixedRateMutation=1.0 and λ divisible by µ → mutations = λ.

---

[arcuri82]

@francastagna hi. this PR has conflicts that need to be resolved before review

[arcuri82]

@francastagna hi. this PR has conflicts that need to be resolved before review