While I think there are some good QoL updates here (the updates to the invocation options), there is something fundamentally wrong with how this seems to attempt to plot data that likely has at least something to do with the index out of bounds error line. It is using contact.id.id-1 instead of just contact.id.id, making the assumption that entity_id will never be lower than 1, and then shifting all entity ids down by 1 and using them as the index of an array. Some simulation outputs I was testing with were causing an index out of bounds error because of this assumption.

The fix for this index out of bounds error in the PR is a band-aid, not a real robust solution. It should probably be replaced with a map(), or have a more in-depth reimplementation, or maybe even just re-evaluated for overall usefulness.

Because this was producing plots that weren't right, instead of fixing this utility more completely, I instead used it as an example of how to interact with scrimmage's python package for parsing frames.bin, and incorporated a new frames.bin translator into my own project's post processor tool, which is producing correct-looking plots from the frames.bin data.

I'm on the fence about what to do about this one:

• revert the fix conidx = min(con.id.id - 1, 0) back to the original conidx = con.id.id - 1 and then merge in the PR to get all the QoL updates
• just decline this PR and move on

I decided to go ahead and remove the one-line fix for the index out of bounds error I was getting. The only thing remaining in the PR are the QoL updates. This PR can be merged if the QoL updates are desireable.