Epic: LIBERO evaluation examples — tutorial + batch runner

[cagataycali]

Umbrella for LIBERO evaluation example scripts that demonstrate Simulation.evaluate_benchmark on the default MuJoCo backend.

Scope

flowchart LR
  TUT[examples/libero_eval.py<br/>Single-task tutorial] --> RUN
  BATCH[examples/libero_batch_eval.py<br/>Multi-seed runner] --> RUN
  RUN[evaluate_benchmark via MuJoCo]

Loading

Tasks

• Add examples/libero_eval.py demonstrating LiberoAdapter on MuJoCo (examples: add examples/libero_eval.py demonstrating LiberoAdapter on MuJoCo #134)
• Port batch LIBERO experiment runner — examples/libero_batch_eval.py over evaluate_benchmark (examples: port batch LIBERO experiment runner (Python over evaluate_benchmark) #135)
• Cross-link both from README "Simulation (MuJoCo)" section
• Reference from strands-labs/robots-sim MIGRATION.md

Dependencies

• Blocked on PR feat: LIBERO benchmark adapter + BDDL parser (#110) #130 (load_libero_suite + evaluate_benchmark) — landed.

Replaces

#134, #135