[Feature] Add a Genetic-Algorithm-based solver

[Pabloo22]

This algorithm will be implemented within the metaheuristics package, likely employing a similar approach to simulated annealing. Comment on this issue discussing your suggested implementation if you would like to contribute.

[Akhilanandateja]

I am on it

[Pabloo22]

Hi Akhilanandateja,

Thanks for your interest in contributing! I currently don't have the time to provide extensive directions as I usually do in good first issues, but I can point you to the requirements of this issue:

• Follow the CONTRIBUTING.md guide.
• Follow the conventions of this library. I recommend reading other source files and ensuring you follow the same style and conventions. Be very attentive to detail.
• Explain your design decisions here. This can save you time by making sure your idea will be accepted before working on it.
• Use a third-party library for the genetic algorithm logic. I'm unsure how to ensure your implementation is correct otherwise. This will also save you time and will probably improve the quality of the implementation. I'm open to being convinced if you think otherwise, though.
• Being able to customize as many elements of the process as possible is essential. There are many possible chromosome representations, for example. We should use a design pattern such as the plugin pattern or dependency injection. See the SimulatedAnnealingSolver. In that solver, we pass functions that compute both the energy and a solution's neighbor, allowing users to come up with arbitrary implementations of these.
• There is literature about genetic algorithms for solving the JSSP already; we should try to implement the strategy of one of these papers. Ideally, we want to support release dates, due dates, deadlines, and flexible variants.
• We also need a tutorial explaining how it works.

I saw you already opened a PR. I hope this also serves as a review. Let me know if you have any questions and if you're still interested in working on this.