Record proxy request metrics

[andrew]

Connects the existing proxy_requests_total and proxy_request_duration_seconds metrics to the request middleware.

Package routes receive canonical labels, including rubygems, golang, packagist, and oci. Non-package routes use other, while /metrics is excluded so scrapes do not count themselves. The middleware shares one duration measurement with request logs and access-log records.

This is stacked on #269 because both changes touch LoggerMiddleware.

[Copilot]

Pull request overview

Connects existing Prometheus request metrics to the server’s request logging middleware so request counts and durations are recorded per request with consistent timing across logs, access-log entries, and metrics.

Changes:

• Record proxy_requests_total and proxy_request_duration_seconds in LoggerMiddleware, excluding /metrics.
• Add path-to-ecosystem canonicalization (rubygems, golang, packagist, oci, else other) for metric labels.
• Add unit tests covering metric recording, /metrics exclusion, and ecosystem mapping.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
README.md	Documents the two request metrics and their labels in the Prometheus metrics table.
internal/server/middleware.go	Records request count + duration metrics in LoggerMiddleware and adds ecosystem label mapping.
internal/server/middleware_test.go	Adds tests validating request metric increments, /metrics exclusion, and ecosystem labeling.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.