Subtle Discrepancies in PBR Implementation

[yaRnMcDonuts]

The purpose of this issue is to keep track of potential issues that may be causing JME's final PBR render output to not match the Khronos standard.

I have personally noticed some PBR models having an extremely subtle foggy white glare to them. And I also recall a jme user years ago who was complaining that their shiny models look "plasticky" and incorrect in JME

Here are some probable issues to keep an eye on:

1. This PR that addresses a potential math issue in the fZero and metallic calculation: Fix incorrect metallic calculations in PBRLighting.frag #2334
2. JME's MikktSpaceTangentGenerator has many "todo" comments left in from 2016 by the original author (who is no longer around to maintain things) that may indicate issues in the functionality. There have been no apparent major issues in jme's tangents, however there is still the possibility that jme's tangents don't match the industry standard for MikktSpace which could effect the final rendering of PBR models with normal maps. (https://hub.jmonkeyengine.org/t/mikktspacetangentgenerator-fails-with-npe/48114/9)

Please comment here if you think that you may have found any other issues with jme's PBR implementation, or if you are knowledgeable about any of the items currently in the list. Thanks.