Add changelog fragments for golden image visualizer test fixes

Author: matthewtrepte

source/isaaclab_newton/changelog.d/mtrepte-golden-image-visualizer-checks.rst

@@ -0,0 +1,8 @@
+Fixed
+^^^^^
+
+* Fixed stale ``_scene_data_mapping`` in :meth:`~isaaclab_newton.physics.NewtonManager.update_visualization_state`
+  being reused after the visualization model was rebuilt for a stage with a different body count
+  (e.g. switching from a 4-env tiled capture to a 1-env viewport capture within the same process).
+  The mapping is now invalidated when its length does not match the current model's body count,
+  preventing wrong body transforms from being written into the shadow ``state_0``.

source/isaaclab_visualizers/changelog.d/mtrepte-golden-image-visualizer-checks-contamination-fix.rst

@@ -0,0 +1,12 @@
+Fixed
+^^^^^
+
+* Fixed cross-test contamination in golden image tests when tiled-camera and viewport captures
+  run sequentially in the same process.  Three sources were addressed: (1) the stale
+  :class:`~isaaclab_newton.physics.NewtonManager` shadow model from the tiled stage persisting
+  into the viewport test on the PhysX backend (cleared in the between-test prepare step);
+  (2) CUDA RNG state drift causing the initial cartpole pole angle to differ between isolated
+  and suite runs (seed is now applied immediately before ``env.reset()``); and (3) test ordering
+  in both golden test files reordered to run tiled captures before viewport captures to prevent
+  RTX render-product state from contaminating tiled camera output.  The ``newton-kit-viewport``
+  golden image was regenerated to match the corrected deterministic initial physics state.