Add `NewtonMassAPI` with inertia tensor

[andrewkaufman]

Is this a new feature, an improvement, or a change to existing functionality?

New Feature

Feature Description

NewtonMassAPI is an extension to UsdPhysicsMassAPI to allow specifying the 6 valued inertia tensor rather than diagonal intertia & the principal axes.

The main value adds are:

• Avoids eigenvalue decomposition for source data which already represents the tensor values
• Its 6 values (whereas MassAPI requires 7)
• Will be double precesion to avoid lossy conversion

We will use -inf as a sentinal value to indicate "no opinion"

If there is an opinion, physics:diagonalInertia and physics:principalAxes should be ignored. ie. this attribute overrides the others when both are authored.

We need validators to ensure values are in a sensible range.

Additional context or alternatives

Open Questions:

• What's the name? newton:localInertia? newton:intertia? newton:inertiaTensor?
• Do we use double[] or GfMatrix3d?
  • There is no Vec6d type so we need to pick between these. Both require validation.
  • double[] requires more documentation about max length (6 values only) & ordering convention
  • GfMatrix3d requires validation that its a semmetric upper triangular matrix

Code of Conduct

• I agree to follow newton-usd-schemas Code of Conduct
• I have searched the open issues and have found no duplicates

[andrewkaufman]

@vastsoun @preist-nvidia @Milad-Rakhsha-NV @havess does anyone have thoughts on the open questions in the description?

[vastsoun]

1. In terms of naming, I would opt for newton:intertia, since this keeps it simple and consistent with newton:mass.

2. Regarding the type, I think double[] seems the lesser of the two evils, as authoring and checking 9 values + symmetry vs 6 seems easier and less error prone + takes up less memory.