Separate out standard of protection (SoP) levels from flood models, adding SoP as a separate hazard indicator

[joemoorhouse]

On on-boarding the TU Delft flood models and commercial models (code not yet in physrisk main) we have integrated flood standard of protection in the flood hazard indicators. This was probably a bad move for 3 reasons (!):

1. It is desirable to make standard of protection and unprotected flood depth arbitrarily combinable (easier to compare different flood models)
2. SoP applies to current conditions
3. Exactly how unprotected flood curves are changed by SoP should be a modelling choice and something that can be modified

This change (to hazard and physrisk) is to make SoP a hazard indicator in its own right. This splits the TU Delft sets into two and adds SoP into physrisk explicitly.