[bug-fix] Object-attached Agent with Split Semantic Scene

[aclegg3]

Motivation and Context

Python implementation of split SEMANTIC render assigns the agent SceneNode's parent to the semantic SceneGraph root before rendering. Attaching an agent to a simulation object creates all rendering assets for the object under the agent's SceneGraph sub-tree. This results in an error during the secondary SEMANTIC rendering pass over the active SceneGraph added in #679.

For reference this error displays as:

SceneGraph::Object::transformations(): the objects are not part of the same tree

This PR re-assigns the agent SceneNode's parent to the active SceneGraph root before the 2nd pass to fix this bug.

How Has This Been Tested

Still no separate semantic testing on CI, leading to this bug slipping through. Tested locally.

Types of changes

• Docs change / refactoring / dependency upgrade
• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist

• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have completed my CLA (see CONTRIBUTING)
• I have added tests to cover my changes.
• All new and existing tests passed.

[codecov]

Codecov Report

Merging #694 into master will increase coverage by 2.09%.
The diff coverage is 0.00%.

@@            Coverage Diff             @@
##           master     #694      +/-   ##
==========================================
+ Coverage   59.87%   61.96%   +2.09%     
==========================================
  Files         157      149       -8     
  Lines        7993     7139     -854     
  Branches       84       84              
==========================================
- Hits         4786     4424     -362     
+ Misses       3207     2715     -492     

Flag	Coverage Δ
#CPP	58.13% <ø> (+2.86%)	⬆️
#JavaScript	10.00% <ø> (ø)
#Python	79.37% <0.00%> (-0.03%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
habitat_sim/simulator.py	95.23% <0.00%> (-0.42%)	⬇️
src/esp/core/random.h	0.00% <0.00%> (-50.00%)	⬇️
src/esp/physics/RigidScene.h	14.28% <0.00%> (-35.72%)	⬇️
src/esp/physics/RigidBase.h	16.07% <0.00%> (-11.90%)	⬇️
src/esp/physics/RigidObject.h	75.00% <0.00%> (-10.72%)	⬇️
src/esp/scene/SceneNode.h	83.33% <0.00%> (-8.34%)	⬇️
src/esp/sim/Simulator.h	80.00% <0.00%> (-5.72%)	⬇️
src/esp/scene/ObjectControls.h	0.00% <0.00%> (ø)
src/esp/physics/RigidScene.cpp
src/esp/agent/Agent.cpp
... and 13 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b3cb98e...882f87b. Read the comment docs.

[Skylion007]

Oof, that 0.0% of the diff hit just underscores the problem. Hope we can get a semantic scene for CircleCI!