Allow resetting model matrix when tiles have regions

[javagl]

Description

When loading a tileset where the tiles contained bounding regions, changing the modelMatrix of the tileset, and then setting the modelMatrix back to its initial value, the update caused a

TypeError: result.computeBoundingVolumes is not a function

A detailed description of what went wrong is in #12002 (comment)

While creating a (pretty strict, dedicated) spec, I noticed that the suggested fix does not fully cover it. There are basically four cases to consider:

• When the transform did change (compared to the initial one)
  • When the result is a TileOrientedBoundingBox, then reuse it
  • When the result is not a TileOrientedBoundingBox, then create a new one
• When the transform did not change (compared to the initial one)
  • When the result is a TileBoundingRegion, then reuse it
  • When the result is not a TileBoundingRegion, then create a new one

(It's hard to be 100% sure that all cases can happen in practice, but easy to handle them generically)

Issue number and link

Fixes #12002
Fixes #12403

Testing plan

The issue descriptions contain examples that caused the crash (basically the same example, though).

Also, I added a dedicated spec for this.

NOTE: The whole spec is currently wrapped into a expect ... not.toThrow. This is the actual change that is done here. One could consider this as the "baseline", though, and remove this check, and only use the expect's for the respective bounding volume types.

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, @javagl!

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

[jjhembd]

@javagl thanks for the fix! The code change makes sense, and the Sandcastle now works for me. I am still trying to run the specs--need to fix a configuration somewhere on my new laptop.

Can you add a note to CHANGES.md?

[jjhembd]

I fixed my configuration and the tests ran through successfully.

Thanks again @javagl!