Feat: Added Mantis Shrimp Optimization Algorithm (MShOA)

[gunbaz]

🚀 Feat: Mantis Shrimp Optimization Algorithm (MShOA) Implementation

📝 Overview

This PR introduces a robust and fully vectorized implementation of the Mantis Shrimp Optimization Algorithm (MShOA), a novel bio-inspired meta-heuristic based on the hunting mechanisms of the mantis shrimp.

The algorithm has been rigorously validated against classic Unimodal and Multimodal benchmark functions with 30 Dimensions and 30 Runs to ensure statistical reliability.

✨ Key Features

• Vectorized Performance: Optimized using NumPy for high-speed computation.
• Statistical Validation: All benchmarks conducted with 30 independent runs.
• Global Optimum Success: Achieved theoretical global optimum (0.0) on complex functions like Rastrigin, Griewank, and Sphere.
• Standardized Logging: Integrated detailed logging for experimental analysis.

📊 Benchmark Results (30D - 30 Runs)

Function	Dim	Runs	Best	Mean	Std. Dev
Ackley	30	30	4.440892e-16	4.440892e-16	0.0
Griewank	30	30	0.0	0.0	0.0
Rastrigin	30	30	0.0	0.0	0.0
Rosenbrock	30	30	28.88	28.96	0.029
Schwefel 2.20	30	30	2.59e-30	7.61e-28	1.01e-27
Schwefel 2.22	30	30	2.59e-30	6.16e-28	8.68e-28
Sphere	30	30	1.59e-61	1.87e-56	4.27e-56
CEC 2017 F1	30	10	5.65e+10	6.85e+10	7.05e+09

🧪 Validation Details

• Environment: Python 3.11
• Libraries: numpy, opfunu (for CEC benchmarks)
• Stability: Demonstrated high convergence speed on unimodal functions and strong exploration capability on multimodal landscapes.

Looking forward to your review!