MycoNet: A Python Framework for Mycorrhizal Network Biophysics — Algorithms, Simulation, and Validation of the Freiman–Villani Efficiency Bounds

MycoNet is an open-source Python package for simulating mycorrhizal network biophysics within the Freiman-Villani thermodynamic framework. The package implements four core algorithms: (1) the local Freiman index via k-nearest-neighbour sampling and hex-integer FFT Minkowski sum, recovering K_hex = 19/7 exactly on the hexagonal lattice; (2) stochastic hyphal growth with Fokker-Planck nutrient transport; (3) Wasserstein W2 distance via log-domain Sinkhorn (POT); and (4) metabolic dissipation via Fisher information. Under simulated drought stress, sigma_r rises from 3.1 to 4.6 and dissipation increases by factor ~24, consistent with the Theorem 6.1 lower bound Psi >= C* D eps^{-2} (sigma_r - K_hex)^2 with C_sim = 20.9 vs C* = 21.8 (4% agreement). The companion theory paper is submitted to the Journal of Mathematical Biology.