Add run statistics and summary output

[Copilot]

Displays formatted statistics at the end of each run: model timing/tokens, action counts, and per-issue operation summary.

Changes

• Statistics tracking (src/stats.ts): Collects model run metrics (duration, p95, token usage), action counts (triaged/skipped/failed), and API call totals
• Token usage extraction (src/gemini.ts): Returns promptTokenCount and candidatesTokenCount from response metadata
• API call counting (src/github.ts): Increments counter on each API method, exposed via getApiCallCount()
• Operation details (src/triage.ts): Each operation implements getActionDetails() for summary formatting
• Integration (src/index.ts): Tracks timing around model calls, records operations, prints summary before exit

Output

📊 Run Statistics:
  Fast
    Total: 1m12s • Avg: 1.7s • p95: 3.4s
    Tokens used: 39k input, 10k output
  Pro
    Total: 3m32s • Avg: 5.7s • p95: 7.4s
    Tokens used: 4.0k input, 1.0k output
  Actions performed: ✅ 38 triaged, ℹ️ 59 skipped, ❌ 1 failed
  GitHub API calls: 199

📋 Summary of Actions Performed:
  owner/repo#42: labels: +bug, +enhancement
  owner/repo#7: comment
  owner/repo#15: title change
  owner/repo#99: state: completed

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Add summary/statistics to the end of run</issue_title>
<issue_description>Log useful results at the end of the run. Use colors, emojis, and friendly formatting. Add tests.

Example (include more than just these):

📊 Run Statistics:
  Fast
    Total: 1m12s • Avg: 1.7s • p95: 3.4s
    Tokens used: 39k input, 10k output
  Pro
    Total: 3m32s • Avg: 5.7s • p95: 7.4s
    Tokens used: 4k input, 1k output
  Actions performed: ✅ 38 triaged, ℹ️ 59 skipped, ❌ 1 failed
  GitHub API calls: 199

📋 Summary of Actions Performed:
  danielchalmers/AutoTriage#42: labels: +bug, +enhancement
  danielchalmers/AutoTriage#7: comment
  danielchalmers/AutoTriage#15: title change
  danielchalmers/AutoTriage#99: state: completed

</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Add summary/statistics to the end of run #26

---

💡 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.