Persistence of Results

[meatballs]

We should create a model(s) to allow us to persist tournament results in the database.

This will allow the data produced to be made available to anyone who wishes to analyse it.

[nealedj]

The results are already persisted in the tournament table. I'm not sure if this is enough though. Does this need augmenting?

[meatballs]

I created this issue simply as an attempt to record the discussion early on Thursday morning at Djangocon where, I think, we concluded that we needed models for the results themselves.

The use case was a researcher who perhaps wanted to study how machine learning strategies perform across all tournaments.

We also recognised that this was a longer term development.