OIT and log depth slowness fix

[lilleyse]

Fixes #7018

OIT was checking if log depth was enabled by checking if the derived command exists. Now it checks if frameState.useLogDepth is true which is more reliable.

This worked before because the log depth command was set to undefined if log depth was off (fb63af). This was changed to make it more efficient to call the pick from ray functions which if called often will constantly change the log depth state, where it's better to keep the log depth commands around.

@bagnell probably best for you to review since it's a follow up to #6957

[cesium-concierge]

Thanks for the pull request @lilleyse!

• ✔️ Signed CLA found.
• ❔ CHANGES.md was not updated.
  • If this change updates the public API in any way, please add a bullet point to CHANGES.md.

Reviewers, don't forget to make sure that:

• Cesium Viewer works.
• Works in 2D/CV.
• Works (or fails gracefully) in IE11.

---

I am a bot who helps you make Cesium awesome! Contributions to my configuration are welcome.

🌍 🌎 🌏

[bagnell]

I see ~40 test failures.

[lilleyse]

@bagnell thanks for catching that, I should have ran the tests locally.

The fix was to not look for derived log depth commands on the inverse classification commands. The classification related sandcastle demos look good now.

[bagnell]

The fix was to not look for derived log depth commands on the inverse classification commands. The classification related sandcastle demos look good now.

That was probably a copy-paste error since they're a sort of post-process.

Looks good 👍