We want to take some variate (e.g. mass) and fit a distribution that depends on another one (e.g. spin) so that we end up with a fit that can tell us, e.g. if the masses of BHs depend on the spins. To evolve one variate with another, we basically condition one probability on the other (condition mass on spin). To do this, we need both individual distributions to be normalized

• normalize distributions that can be normalized analytically
• normalize others numerically? or just add logic that only allows for conditioning to be done on dists that have flag normalized=True
• add plumbing that is opaque to me at this time