feat(go): add gated OTLP metrics, reconcile events, and log bridge (#2148)

[MartinForReal]

Implements the Go-side observability gaps from #2148 as three independent, default-OFF additions so the Go runtime/controller expose the same OpenTelemetry signals as the Python runtime.

1. Go ADK — GenAI token-usage metrics (go/adk)

Configures a MeterProvider with a periodic OTLP metric exporter and emits gen_ai.client.token.usage (Int64 histogram, unit {token}) per LLM call.

• Attribute set matches the Python runtime so one dashboard covers both runtimes: gen_ai.token.type (input/output), gen_ai.request.model, gen_ai.response.model, gen_ai.provider.name, gen_ai.agent.name.
• The A2A executor records one input + one output observation per LLM call (prompt tokens; output = candidate + reasoning tokens), skipping streamed Partial events so chunked streams are not double-counted.
• Provider/model labels resolved from agent config (incl. config_usage.ModelName); response model falls back to the request model.
• Gated behind OTEL_METRICS_ENABLED (default OFF) and the standard OTEL_EXPORTER_OTLP_* endpoint/protocol resolution shared with traces.
• Tests: adk/pkg/telemetry/metrics_test.go (recording via ManualReader, provider-name mapping) and adk/pkg/a2a/executor_metrics_test.go (token accounting + partial/nil guard).

2. Kubernetes Events on reconcile (go/core)

The discovery reconcilers surface lifecycle events via an optional EventRecorder, mirroring the agentharness-substrate-controller pattern:

• RemoteMCPServer (internal/controller/remotemcpserver): Normal ToolsDiscovered, Warning ValidationFailed, Warning ReconcileFailed.
• MCPServer (MCPServerTool role, internal/controller/mcpserver): Normal ToolsDiscovered, Warning ValidationFailed, Warning ReconcileFailed.

Note on Agent/ModelConfig: in current main those objects reconcile through the KRT-based status collections in go/core/v2/controller, not classic controller-runtime Reconcile installers, so they have no EventRecorder path to attach events to. The events gap is therefore implemented on the two remaining classic discovery controllers — the direct successors of the MCPServerTool/RemoteMCPServer roles named in the issue. Emission is a no-op when no recorder is wired (the default).

3. Controller log → OTLP bridge (go/core)

• telemetry.InitLoggerProvider builds an OTLP LoggerProvider via autoexport (same resource attributes as traces) and sets it global.
• telemetry.ControllerZapOpts additively tees the controller zap logger with an otelzap bridge core, preserving stdout.
• Wired into cmd/controller-v2/main.go before the controller logger is built.
• Gated behind OTEL_LOGGING_ENABLED (already plumbed via the Helm chart otel.logging.enabled); when unset both functions are no-ops and logs are byte-identical.

Validation

• go build ./..., go vet on all changed packages, gofmt/goimports clean.
• golangci-lint on changed packages: 0 issues.
• go test passes for adk/pkg/a2a, adk/pkg/telemetry, adk/pkg/config, core/internal/telemetry, and core/internal/controller/{mcpserver,remotemcpserver} (incl. the new event tests).
  • pkg/session has two pre-existing Windows-only TempDir-cleanup failures (SQLite handle held open during cleanup); unrelated to this change.

Dependencies

Promotes otlplog, otlpmetric(grpc/http), otel/log, otel/metric, otel/sdk/metric to direct requires at the same versions already in the graph, and adds go.opentelemetry.io/contrib/bridges/otelzap v0.19.0. No version bumps.

Refs: #2148

[MartinForReal]

Hi @supreme-gg-gg, friendly ping for review when you have a moment. This is the gated Go observability work for #2148 (OTLP metrics, reconcile Events, controller log to OTLP bridge). I have rebased it on latest main and resolved the conflicts, so it is ready for another look. Thanks!

[MartinForReal]

Addressed: the adk-side OTLP token metric that duplicated #2673 was removed entirely (go/adk/pkg/telemetry/metrics.go and its hooks/tests are gone from this branch). The PR now stays focused on the controller-scoped OTLP metrics, reconcile events, and the controller log bridge; the GenAI token metric lives only in #2673.

[github-actions]

This pull request has been marked as stale because of no activity in the last 15 days. It will be closed in the next 5 days unless it is tagged "no stalebot" or other activity occurs.