Reduce memory and disk-space usage of `HypoFitResult`

[thehrh]

• an analyser making use of https://github.com/icecube/pisa/blob/dacc631a04821003aa012e74908bda689c5936c3/pisa/analysis/analysis.py#L963 is unable to switch off recording and storing of fit/minimiser history, best-fit asimov template, detailed metric maps

• depending on nature of analysis, this can give rise to huge memory footprint, and, if no manual purging is done before storing, may take up unnecessary disk space

• unfortunately, there is little documentation explaining the design choices behind the HypoFitResult class and its integration into the fitting functions (e.g., it's not obvious to me that the deserialisation mechanism should require all attributes in _state_attrs to be present)

• setting blindness > 0 suppresses storing fit_history, but would have other unintended side effects for an analyser only interested in reducing analysis resource usage

• introduce a parameter to the call to fit_recursively which will force the fit to omit the most detailed information?