Use performance optimized shuffle_do() method

[EwoutH]

Replace shuffle().do() in 18 models with the performance optimized shuffle_do() method.

shuffle_do() was added in projectmesa/mesa#2283 and is first included in the upcoming Mesa 3.0.0a5 release. It allows for a drop-in performance upgrade:

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

Which can speed up the Agent activation by over 3x depending on the scenario.

Configuration	shuffle().do()	shuffle(inplace=True).do()	shuffle_do()
10,000 agents, 1,000 steps	8.27 s	3.65 s	3.06 s
100,000 agents, 100 steps	13.71 s	6.31 s	3.71 s
1,000,000 agents, 10 steps	18.36 s	9.44 s	5.75 s

shuffle_do() can be used to replace the old RandomActivation scheduler, as demonstrated in #183.

Part of #111.

[EwoutH]

Thanks for the swift reviews!