Individual model checks & null model

[kathsherratt]

@seabbs comments on outlier forecasts (wider than #85 so separating into an issue):

I assume that most of the crazy is coming from the Rt forecast (or maybe not?) and so it might be better to do model by model checks and then reset crazy forecasts to something less mad or drop them from the ensemble (but that gets quite complicated when weighting as if a model is not present the weighting will no longer add up to 1).

My preference is that we start submitting forecasts for all of the targets always. Given that it would be good if we swapped crazy forecasts with some kind of null model (no change with overdispersion from recent obs - I think @nikosbosse has code for this).
So the pseudo code would be if not present add in null model forecast and otherwise don't?

Need to:

• Implement checks on each model for outlier forecasts that are not close to data (see Check 1 wk ahead forecast is somewhere near data #85 but for each model before ensembling)
• Set up a null model (code from @nikosbosse)
• Replace outlier forecasts with null model in ensemble