Wrap QuinticPathPlanner target yaw changes

[sylvesterkaczmarek]

Description

Fix the remaining yaw-change comparison issue in QuinticPathPlanner.

Isaac Sim 6.1 already contains the upstream fix for signed yaw extraction from targetPrim. The remaining comparison still subtracts yaw values directly, so physically close headings on opposite sides of +pi/-pi can look almost 2*pi apart and trigger an unnecessary replan.

This change compares the normalized angular delta instead, using the existing normalize_angle() helper.

Validation

• regression added to the existing navigation quaternion test suite
• verifies +179° -> -179° is treated as an approximately 2° change
• source and test files pass py_compile
• patch passes git diff --check
• production change is one expression

[sylvesterkaczmarek]

Rebased onto the 6.1 GA develop branch and narrowed to the remaining issue: target-yaw change detection now uses the shortest wrapped angular delta. The signed targetPrim yaw fix is already upstream. Added the regression to the existing navigation test suite; py_compile and git diff --check pass.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.