These changes replace the formula for accuracy pp, as well as penalise rhythm difficulty from pp based on estimated unstable rate.

More effort has been put in to balance this than my other recent changes. The new curve is steeper than the current (making improving your acc on a play more rewarding) and attempts to break even with the current curve at 98% accuracy on a nomod map of the given star rating, based on estimated UR data from a large number of ranked maps. See said data and how the curves compare here https://www.desmos.com/calculator/hwd2pjmqqr

The rhythm penalty uses the same data but for 100% acc rather than 98% to estimate the unstable rate at which all of a map's rhythm difficulty has been played successfully. 2x this unstable rate is around 90-91% on the same maps which is where I think it's safe to say all rhythm difficulty has been ignored. Rhythm difficulty / star rating has been a good measure of rhythm contribution to a map since #32426, with around 0.15 being where rhythm starts to have any effect on star rating and 0.35 being maps like 3772247 (hard rhythm but mechanically easy) and 4543374 (very hard rhythm). Removing all rhythm strain from these 0.35 maps reduces star rating by ~20%, and doing the same on maps in the middle with 0.25 (such as 3658314 and 4725197) reduces star rating by ~5% which is why I went with 0.2 * rhythmFactor^2 for the penalty's max value.

I also renamed the mono acc scaling stuff for more clarity now that we have two acc related scalings in difficulty pp