feat(swarm_based): Add OriginalICSA algorithm and integration tests

[gorkemulas2005]

📑 Description

This PR introduces the Improved Chameleon Swarm Algorithm (ICSA) to the swarm_based module. ICSA is a nature-inspired metaheuristic based on the 2024 paper published in Biomimetics. The implementation focuses on multi-strategy fusion to enhance population diversity and avoid local optima, utilizing fully vectorized NumPy operations for maximum efficiency within the Mealpy framework.

✨ Key Features Implemented

Stage	Strategy	Description	Reference
Initialization	Logistic Chaotic Map	Enhances initial population diversity and ergodic properties.	Eq. 9, 10
Prey Searching	Sub-population Hybrid	Balances exploration via a hybrid spiral search mechanism.	Eq. 11
Eyes' Rotation	Lévy Flight Stage	Adaptive rotation using Lévy flight and cosine factors.	Eq. 15
Prey Capture	Refraction Reverse-Learning	Prevents stagnation by modeling light refraction principles.	Eq. 20

🛠 Implementation Checklist

• Full vectorized ICSA implementation in mealpy/swarm_based/ICSA.py
• Proper export in mealpy/swarm_based/__init__.py and mealpy/__init__.py
• Comprehensive unit tests in tests/swarm_based/test_ICSA.py
• Docstrings and mathematical references added to the class

✅ Checks

• My pull request adheres to the code style of this project
• My code requires changes to the documentation
• I have updated the documentation as required (Docstrings added)
• All the tests have passed

ℹ️ Additional Information

The algorithm's robustness was verified on a 16-core system using two benchmark suites.

1. Classic Benchmark Results (F1 - F18)

The implementation matches the global optima and convergence characteristics of the original article.

Func	Best (Min)	Mean	Std	AvgTime(s)
F1	0.0000e+00	0.0000e+00	0.0000e+00	1.26
F2	0.0000e+00	0.0000e+00	0.0000e+00	1.68
F3	0.0000e+00	0.0000e+00	0.0000e+00	18.21
F4	0.0000e+00	0.0000e+00	0.0000e+00	1.20
F5	2.8950e+01	2.8975e+01	1.1124e-02	4.68
F6	4.2906e+00	6.0516e+00	6.3142e-01	1.40
F7	6.4002e-07	1.0960e-05	1.2816e-05	2.90
F8	-6.6905e+03	-5.5618e+03	5.3885e+02	1.50
F9	0.0000e+00	0.0000e+00	0.0000e+00	2.06
F10	4.4409e-16	4.4409e-16	0.0000e+00	2.50
F11	0.0000e+00	0.0000e+00	0.0000e+00	2.23
F12	2.5722e-01	7.5567e-01	2.7886e-01	7.27
F13	1.9366e+00	2.9157e+00	2.1140e-01	6.46
F14	2.0000e-03	2.0000e-03	5.4857e-19	4.45
F15	3.0847e-04	6.4443e-03	7.8868e-03	3.47
F16	-1.0316e+00	-1.0308e+00	4.3406e-03	0.76
F17	3.9789e-01	3.9789e-01	1.1300e-06	0.89
F18	3.0000e+00	5.7403e+00	8.0895e+00	1.04

2. High-Complexity Benchmarks (CEC 2017 via opfunu)

ICSA performance comparison against Mealpy's PSO on shifted and rotated functions.

Function	ICSA Mean	PSO Mean	ICSA Std	PSO Std	ICSA Time(s)
CEC17_F15	4.3325e+03	2.7233e+03	4.1379e+03	2.6290e+03	15.00
CEC17_F19	4.7720e+03	2.4826e+03	5.4133e+02	3.1292e+02	49.00
CEC17_F20	3.7261e+03	2.5334e+03	1.1981e+03	4.1964e+02	7.64

Verification Result: 1 passed in 1.09s

📚 References

1. Chen, Y.; Cao, L.; Yue, Y. Hybrid Multi-Objective Chameleon Optimization Algorithm Based on Multi-Strategy Fusion and Its Applications. Biomimetics 2024, 9, 583.
2. Malik, B.S. Chameleon Swarm Algorithm: A bio-inspired optimizer for solving engineering design problems. Expert Systems with Applications, 2021, 174, 114685.

---

*Submitted by Kazan Ulaş Görkem