document hawk/dove convergence logic and include converged status in data collection

[rlskoeser]

• update readme to include brief description of convergence logic
• add status to model data collection (converged/running) so it is easy to check if batch runs completed
• see if there's an easy way to display convergence status in mesa interface (near the steps in the sidebar if possible)
• try to determine typical convergence for the longer + more complicated runs included in our batch runs, so we can set a reasonable upper bound

[rlskoeser]

@LaraBuchak I looked into options for displaying the status of the simulation (running or converged) so you'd know why it stopped. I can't put it where I'd prefer to; could you let me know if the easy option would be useful for you?

I would much prefer to display the status in the sidebar, something like this:

That's not currently supported by the mesa experimental jupyterviz display we're using. I could open a pull request with the approach I used to generate this to get feedback on it (and I probably should do that), but I worry about how complicated the jupyterviz interface is already without adding more.

What I was recommended to do, which is possible with current mesa, is include the text as a block with the other "measures" (charts), like this (look under the blue bar chart):

Would this option be useful?

[rlskoeser]

Discussed with @LaraBuchak at meeting 2024-02-21; this convergence logic is fine for the base hawk/dove game, but it doesn't get at what we care about for the hawk/dove multi risk attitude - what we really want there is equilibrium. Merged in #65 but will revisit by testing new convergence logic for the hawk/dove multi game based on agents not changing on adjustment rounds.