Fix specular reflection artifact in PBR direct lighting

[jjhembd]

Description

When computing the specular contribution from PBR direct lighting, the dot products used in the BRDF need to be clamped to the range [0.0, 1.0]. Our code was missing the clamp, but the artifacts didn't appear until we updated the BRDF in #12063.

This PR adds the clamps in computeDirectSpecularStrength in pbrLighting.glsl.

Rendering results

Before this PR:

After this PR:

Issue number and link

Fixes #12109.

Testing plan

1. Load @javagl 's Sandcastle and verify that the artifacts are gone after this PR.
2. Load the glTF PBR Extensions Sandcastle and view models with "Direct lighting only", rotating them to the side opposite the light source.

Author checklist

• I have submitted a Contributor License Agreement
• I have added my name to CONTRIBUTORS.md
• I have updated CHANGES.md with a short summary of my change
• I have added or updated unit tests to ensure consistent code coverage
• I have updated the inline documentation, and included code examples where relevant
• I have performed a self-review of my code

[github-actions]

Thank you for the pull request, @jjhembd!

✅ We can confirm we have a CLA on file for you.

[javagl]

From the user-facing perspective:
I tried it out with my test sandcastle and the "WaterBottle"+directional light (that you mentioned in the issue), and I don't see the artifacts any more.

From the code perspective:
The artifacts looked like "some division-by-epsilon" or ~"some missing clamping" (we roughly know how these things look like 🙂 ). So the change looks reasonable.

But I don't say that with a "deep understanding". For example, that abs was removed from the NdotV . If that line had been changed to something like
clamp(dot(normal, viewDirection)), -1.0, 1.0);
then it would also look reasonable for me, by quickly skimming over it.

(Or to put it simply: I didn't do the math...)

A ping @bertt for the case that you want to try this out early.

[ggetz]

Looks good to me! Thanks @jjhembd and @javagl!