Unify convex hull limits across generated assets

[skywhite1024]

Description

This PR centralizes convex-hull limits for action-agent and Prompt2Scene assets, then applies the shared defaults consistently across task objects, extra rigid objects, tables, gravity preprocessing, gym export, and scene editing.

The unified defaults are:

• Task objects: 8 hulls
• Extra rigid objects: 1 hull
• Tables: 1 hull
• Preprocessing assets: 16 hulls
• Preprocessing tables: 8 hulls

This removes role-specific hard-coded limits and prevents source configuration values from overriding the generated role policy.

Dependencies: None.

Issue: N/A.

Type of change

• Bug fix (non-breaking change which fixes an issue)
• Enhancement (non-breaking change which improves an existing functionality)
• New feature (non-breaking change which adds functionality)
• Breaking change (existing functionality will not work without user modification)
• Documentation update

Screenshots

Not applicable.

Validation

• conda run -n embodichain black --check --diff --color ./: passed, 679 files unchanged.
• Targeted convex-hull tests: 11 passed.
• Broader action-agent generation test module: 111 passed, 32 failed. The failures concern existing robot-profile, camera, layout, and summary expectations; no modified convex-hull assertion failed.
• Headers, future annotations, and git diff --check: passed.

Checklist

• I have run the black . command to format the code base.
• I have made corresponding changes to the documentation (not required; no public API behavior was added).
• I have added tests that prove my fix is effective.
• Dependencies have been updated, if applicable (no dependency changes).