Fix signed distance to bounding sphere.

[bagnell]

Fixes #6451.

CC @emackey

[cesium-concierge]

@bagnell, thanks for the pull request! Maintainers, we have a signed CLA from @bagnell, so you can review this at any time.

⚠️ I noticed that CHANGES.md has not been updated. If this change updates the public API in any way, fixes a bug, or makes any non-trivial update, please add a bullet point to CHANGES.md and comment on this pull request so we know it was updated. For more info, see the Pull Request Guidelines.

---

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

🌍 🌎 🌏

[emackey]

Confirmed fixed. Thanks!

[emackey]

Will wait for Travis...

[emackey]

Hmm. Are there test failures here?

[bagnell]

Looks like it. I couldn't reproduce it locally though.

[emackey]

Scene/LabelCollection computes bounding sphere in 3D
Expected 111158688.34332076 to equal 4862789.037706432.

I was able to see it in Jasmine locally when running the whole test suite, but not running LabelCollectionSpec in isolation. It could be some bad interaction with other tests?

It's too much of a red flag that this is a bounding-sphere related test fail, and I don't see it in master or other recent PRs. I think it's local to this branch, unfortunately.

[emackey]

Also Travis didn't get the same result as Jasmine. Here's Travis:

Expected 107984155.04338884 to equal 4862789.037706432.

[mramato]

Just to confirm, I can reproduce the failure locally every time with:

npm run test -- --browsers Electron --webgl-stub --failTaskOnError --suppressPassed

it does not happen at all for me in master

[bagnell]

I reverted the changes to what was in 1.44. The distance to bounding sphere wasn't needed for log depth but I noticed that it was returning a size for volumes behind the camera.

[emackey]

Looks good, tests pass. Thanks @bagnell.