Fix medium layer base weight in testrender

[tdavidovicNV]

Restore the pre-rewrite medium traversal behavior by avoiding a second application of the incoming weight to a layer base. Add a render regression comparing directly weighted and layered media.

Assisted-by: OpenAI Codex / GPT-5

Description

The non-recursive process_medium_closure traversal introduced in #1897
applies the incoming weight twice when it saves the base of a MaterialX layer.

base_w already includes the incoming weight:

base_w = weight * (1 - opacity(top));

The iterative walker then saves weight * base_w, giving a layer reached
with weight s an s * s factor on its base medium. The previous recursive
walker passed base_w directly, as does the current BSDF walker.

This change restores that behavior by saving base_w without multiplying by
the incoming weight again.

This is an internal testrender correction with no OSL language or public API
change, so no documentation update is needed.

Tests

The render-mx-medium-vdf test now renders a directly weighted absorbing
medium beside the same medium under a negligible layer top:

0.5 * medium
0.5 * layer(negligible_top, medium)

With the fix, the two spheres match. Before the fix, the layered sphere uses
0.5 * 0.5 for its medium weight and is visibly less dense; 22.7% of pixels
exceed the test's idiff threshold.

Local validation on Windows:

• Release builds of testrender and oslc completed successfully.
• render-mx-medium-vdf.opt passes through the standard test harness.
• The same scene rendered with the pre-fix binary fails the new comparison.
• clang-format 17 and git diff --check pass.
• OptiX validation is left to CI.

Codex was used to audit the iterative traversal against its recursive
predecessor, prepare the regression scene, and draft the fix and tests. I
reviewed the resulting changes and test output.

Checklist

• I have read the guidelines on contributions and code review procedures.
• I have read the Policy on AI Coding Assistants
  and if I used AI coding assistants, I have an Assisted-by: TOOL / MODEL
  line in the pull request description above.
• I have updated the documentation if my PR adds features or changes
  behavior.
• I am sure that this PR's changes are tested in the testsuite.
• I have run and passed the testsuite in CI before submitting the
  PR, by pushing the changes to my fork and seeing that the automated CI
  passed there. (Exceptions: If most tests pass and you can't figure out why
  the remaining ones fail, it's ok to submit the PR and ask for help. Or if
  any failures seem entirely unrelated to your change; sometimes things break
  on the GitHub runners.)
• My code follows the prevailing code style of this project and I
  fixed any problems reported by the clang-format CI test.

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: tdavidovicNV / name: Tomas Davidovic (dc6500d)

[aconty]

Good catch. LGTM!

[tgrant-nv]

Yes, good catch! Thanks for cleaning up my mess.

[lgritz]

I'm seeing that sanitizer failure on multiple branches, so I'm sure it's not related to this PR and will not let that hold things up.

Great fix, merging.