hfloat and html report fixes

[figment]

Add support for hfloat (not IEEE 754-2008 compliant)
Fix html report for python 3

[psi29a]

This would allow for using the latest nif.xml file which included hfloat. :)

[neomonkeus]

Check out my branch here - develop...neomonkeus:hfloat
It has some required additions that I would want to have included along with this.

Regarding usage of latest nif.xml. It breaks in several other areas unfortunately. I think we would need to sync up with @jonwd7. There are probably several other refactors by jon that need to be included into the @niftools/nifxml so need to plan those changes. Ideally we would want a phased nif.xml version officially. i.e add support for hfloat and other types first. then apply usage, with the eventual goal of the niftools version converging back to the version jon uses now for @niftools/NifSkope

[3blake7]

i know this is a work-in-progress, i just wanted to let y'all know that I patched the latest source from git with the commits above and latest nif.xml. i no longer get the hfloat error. i am using niftoaster with the check_centerradius spell, which runs on my nifs without any errrors but the bounding spheres which have vertices outside the shell are not updated and the timestamps on the nifs is unchanged.

[neomonkeus]

@fractalfederation Although this would update add the base support to read the block into memory, the way the BSTriShape stores the vertex data probably would require the spell to be updated also. Will update your ticket to track that as needing to be done.

[JayFoxRox]

What's the status on this?

While I don't care about hfloat, I would still like to use the latest nif.xml.
It's a bit annoying to backport changes / bugfixes.