Fix quality_avg to average final_score instead of computing completion rate

[Copilot]

quality_avg was reporting completion rate (completed traces / total traces × 100) rather than the average quality score from final_score values on each trace.

Changes

• observability.py: Replace the completion-rate formula with a proper average of final_score across completed traces that have a non-None score

# Before — measures completion rate, not quality
"quality_avg": round(
    sum(1 for t in recent_traces if t.status == "completed")
    / max(len(recent_traces), 1)
    * 100,
    1,
),

# After — true average of final_score values
completed_quality_scores = [
    t.final_score
    for t in recent_traces
    if t.status == "completed" and t.final_score is not None
]
quality_avg_value = (
    (sum(completed_quality_scores) / max(len(completed_quality_scores), 1)) * 100
)
"quality_avg": round(quality_avg_value, 1),

Falls back to 0 when no qualifying scores exist via max(len(...), 1).

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.