feat: adding abstraction of optimal agent movement

[adamamer20]

It would be beneficial to have a function in the DiscreteSpaceDF (and AgentContainer) for moving agents based on attributes:

move_to(agents: AgentLike, attr_names: str | list[str], rank_order: ['max' | 'min'] | list['max' | 'min'])

For example: move_to('sugar', 'max'). This would abstract the code for handling sequential conflicts, making it easier than handling them from scratch each time.

[suryanshgargbpgc]

hi @adamamer20, I tried to understand this issue, if I am correct this feature aims to write a function which upon calling would solve the issue of multiple agents trying to move on the same grid at the same time which will prevent us from dealing with those edge cases everytime.
Can you please share me necessary resources which can help me address this issue ?
Also could you please elaborate on any points i missed.

[adamamer20]

@suryanshgargbpgc You're correct in your understanding! I recommend checking out the sugarscape example for polars to get a clear picture of these concurrency conditions. You can see it here:

https://github.com/projectmesa/mesa-frames/blob/1d7b3ed11a61130cbee12fe0c9222d233b234669/examples/sugarscape_ig/ss_polars/agents.py#L174C1-L225C73

In this example, the best move might be the same for multiple agents at the same time. Since we're moving all ants simultaneously in a vectorized way, it's important to address this issue. I handled it this way in sugarscape, and I believe the same reasoning can be generalized to accommodate various agent attribute objectives, removing the burden of the implementation on the user