refactor(api): move metrics initialization into api crate

[poroh]

Reduce the scope and growth of carbide-api-core by moving process initialization into the carbide-api composition crate. This PR moves metrics initialization, metrics endpoint startup, and process task ownership without intended functional or configuration changes.

This is the second change in a planned series:

1. Move logging initialization and the top-level run entry point - refactor(api): move logging initialization into api crate #3851.
2. Move metrics initialization, the metrics endpoint, and process task ownership - this PR.
3. Move dynamic settings, database, Vault, credentials, certificates, KMS, and secret-import setup.
4. Move IPMI, Redfish, RMS, InfiniBand, NMX-C, DPF, and component-manager client setup.

The existing metric views and gauge regression coverage are preserved. Metrics-bootstrap dependencies that are no longer used by api-core move with the implementation.

Related issues

#3851

Type of Change

• Add - New feature or capability
• Change - Changes in existing functionality
• Fix - Bug fixes
• Remove - Removed features or deprecated functionality
• Internal - Internal changes (refactoring, tests, docs, etc.)

Breaking Changes

• This PR contains breaking changes

Testing

• Unit tests added/updated
• Integration tests added/updated
• Manual testing performed
• No testing required (docs, internal refactor, etc.)

Additional Notes

This PR is stacked on #3851 and should merge after it.

[copy-pr-bot]

Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[coderabbitai]

Walkthrough

API metrics ownership moves into crates/api, where Prometheus/OpenTelemetry setup and the metrics endpoint are initialized. api-core consumes caller-provided telemetry and task orchestration through CoreRunInputs, while obsolete metrics dependencies and setup code are removed.

Changes

API metrics lifecycle

Layer / File(s)	Summary
Metrics provider and endpoint setup crates/api/Cargo.toml, crates/api/src/lib.rs, crates/api/src/metrics.rs, crates/api-core/src/bootstrap.rs, crates/api-core/src/logging/setup.rs, crates/api-core/Cargo.toml	The API crate adds Prometheus-backed OpenTelemetry metrics, histogram views, span-count gauges, and validation coverage; obsolete metrics setup and dependencies are removed from api-core, and ApiMetricsEmitter is re-exported.
Core runtime telemetry and task injection crates/api-core/src/run.rs	CoreRunInputs accepts a meter and borrowed JoinSet; run_core passes them through runtime initialization and no longer owns metrics creation or final task waiting.
Application task lifecycle wiring crates/api/src/run.rs	run initializes metrics, starts the optional metrics endpoint, passes shared runtime state to run_core, and waits for spawned tasks through join_all.

Estimated code review effort: 4 (Complex) | ~45 minutes

Sequence Diagram(s)

sequenceDiagram
  participant api_run as api::run::run
  participant metrics as api::metrics::setup_metrics
  participant endpoint as start_metrics_endpoint
  participant core as api_core::run_core
  participant tasks as Tokio JoinSet
  api_run->>metrics: initialize Prometheus/OpenTelemetry metrics
  metrics-->>api_run: return registry and meter
  api_run->>endpoint: spawn optional metrics server
  api_run->>core: provide meter and shared JoinSet
  core-->>api_run: return after API startup
  api_run->>tasks: wait with join_all

Loading

Possibly related PRs

• NVIDIA/infra-controller#3851: Both changes modify logging setup in crates/api-core, including removal of metrics and tracing-related wiring.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 66.67% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Title check	✅ Passed	The title accurately summarizes the main refactor: moving metrics initialization into the api crate.
Description check	✅ Passed	The description matches the changeset and describes the metrics refactor, endpoint startup, and task ownership move.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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