docs(proxy): document --prometheus_metrics_port separate metrics server

[devin-ai-integration]

Summary

Documents the new --prometheus_metrics_port / PROMETHEUS_METRICS_PORT option added in BerriAI/litellm#39889. Adds a "Serve /metrics from a separate process" subsection to docs/proxy/prometheus.md under the multi-worker section: why you would want scrapes off the inference workers, the startup command, the /metrics and /health endpoints on the metrics port, a Prometheus scrape_configs target, and the operational notes (same --host, needs the prometheus callback, exits with the proxy, multiproc dir auto-created, output identical to the main endpoint which stays mounted, no virtual-key auth on that port, Docker -p 4001:4001, Helm extraEnvVars plus pod-level scraping since the chart Service only exposes the proxy port).

Resolves LIT-6881 (docs part).

Validation: npm run build succeeded locally ([SUCCESS] Generated static files in "build"). Rendered-page screenshot will be posted as a PR comment.

Link to Devin session: https://app.devin.ai/sessions/2f6b60305e844c1387662eb8bda72ff4
Open in Devin Desktop: https://app.devin.ai/desktop/session/2f6b60305e844c1387662eb8bda72ff4?variant=devin
Requested by: @yassin-berriai

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
litellm	Ready	Preview	Sep 5, 2026 9:43am UTC

[devin-ai-integration]

Rendered page from the local npm run build output, new subsection under Multiple Workers