[codex] Track rejected rollout failure metrics

[rasdani]

Summary

• Keep compatibility errored_rollouts/{env} and errored_rollouts/all metrics.
• Add scheduler-side rejected-rollout error metrics derived from rollout["failure"], with legacy fallback classification from rollout["error"].
• Emit error/all/mean, error/{env}/mean, and reason/origin metrics under error_reason/... and error_origin/....
• Count scheduler task exceptions as env_server_error failures instead of only logging Rollout failed: ....
• Preserve accepted-rollout metric aggregation under the existing metrics/{env}/... path; rejected rollout failures are tracked scheduler-side before rescheduling.

Companion verifiers failure-payload PR: PrimeIntellect-ai/verifiers#1283.

Compatibility Note

• The companion verifiers PR intentionally broadens agent_error to all agent-origin failures.
• agent_nonzero_exit preserves the previous narrow non-zero process exit signal.

Validation

• uv run ruff check src/prime_rl/orchestrator/scheduler.py tests/unit/orchestrator/test_scheduler.py
• python -m py_compile src/prime_rl/orchestrator/scheduler.py
• uv run pytest tests/unit/orchestrator/test_scheduler.py