Replace RandomActivation and SimultaneousActivation schedulers with AgentSet functionality

[EwoutH]

This PR replace all the RandomActivation and SimultaneousActivation schedulers used in the example models with AgentSet functionality. This simplifies models and allows for easier and more flexible modification of these models. It algins the examples with the current best practices in Mesa.

RandomActivation (b7f67c7)

The RandomActivation schedulers were removed, and in the Model.step() function the schedule step was replaced with an AgentSet operation, like:

-        self.schedule.step()
+        self.agents.shuffle().do("step")

The following RandomActivation schedulers were removed:

• bank_reserves
• boid_flockers
• boltzmann_wealth_model
• boltzmann_wealth_model_experimental
• boltzmann_wealth_model_network
• caching_and_replay
• charts
• el_farol
• epstein_civil_violence
• forest_fire
• hotelling_law
• schelling
• shape_example
• virus_on_network

SimultaneousActivation (7e396fe)

To remove the SimultaneousActivation schedulers, the Agent methods were renamed from step() and advance() to more descriptive functions. The model step was updated accordingly:

-        self.schedule.step()
+        self.agents.do("determine_state")
+        self.agents.do("assume_state")

The following SimultaneousActivation schedulers were removed:

• hex_snowflake
• color_patches
• conways_game_of_life

Relevant tickets

• Use new AgentSet functionality in examples #111
• Deprecate all schedulers and be free to do whatever you want mesa#1912
• Introduce AgentSet class mesa#1916

[EwoutH]

And believe it or not, the final error is the DataCollector that tries to gather attributes from agents that don't have said attributes.

[EwoutH]

I updated the PR description and title and cleaned up the commit history as far as feasible.

Looking forward to reviews!

[rht]

Other than the 2 comments, everything else LGTM. Feel free to rebase and merge after they had been addressed.

[EwoutH]

With rht’s approval I’m going to merge this.

@jackiekazil and @tpike3 I would still love your take, if you have any feedback let me know.