rate helper quoteError() scale

[pcaspers]

The rate helper implementations use different quote scales. For example, FutureRateHelper uses a quote in the order of 1E2 while SwapRateHelper uses 1E-2.

This becomes an issue when we run a global bootstrap and do not aim for an exact match of the instruments but want to balance the instrument fit with the smoothness of the curve.

Even for simpler setups where we use IterativeBootstrap for an exact instrument match, it would be desirable if all quote errors are in the same order so that the given error tolerance for the solver has a consistent interpretation.

One possible solution would be to add a scale() function to BootstrapHelper that can be used in the bootstrappers to normalize the error.

Thoughts?