feat: expose generic cluster node metrics

[pratik50]

What

Adds active and inactive cluster node counts to the Prometheus registry using generic metrics with a node_type label:

• parseable_active_nodes{node_type="ingestor"}
• parseable_active_nodes{node_type="querier"}
• parseable_inactive_nodes{node_type="ingestor"}
• parseable_inactive_nodes{node_type="querier"}

Node status is determined using the existing liveness check.

Why

Using generic metrics avoids separate metric definitions for every node type and allows analytics to filter or group node counts by node_type.

Validation

• cargo fmt --all -- --check
• cargo check --lib --bin parseable
• cargo clippy --lib --bin parseable -- -D warnings

[coderabbitai]

Navigate logical layers of code changes, visualize relationships, and explore their blast radius.

Note

Currently processing new changes in this PR. This may take a few minutes, please wait...

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Essentials

Run ID: 31b933af-0a6f-484e-9c0e-3e7fc11d3668

📥 Commits

Reviewing files that changed from the base of the PR and between b09ab6f and 9e19567.

📒 Files selected for processing (2)

• src/analytics.rs
• src/metrics/mod.rs

 ________________________________________________________________________
< OpenAI said I could be anything I wanted, so I became a code reviewer. >
 ------------------------------------------------------------------------
  \
   \   (\__/)
       (•ㅅ•)
       / 　 づ

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create a new PR

---

_{Comment @coderabbitai help to get the list of available commands.}

[pratik50]

Closing this PR because the change belongs in the existing analytic-node-counts branch.