Add OTEL K8s events integration tests

[louisall]

Description of the issue

Adds integration test coverage for the K8s events pipeline that was added in:

• feat(nodemetadataenricher): Add log signal support amazon-cloudwatch-agent#2095 (nodemetadataenricher log signal support)
• feat: Register k8seventsreceiver in default components amazon-cloudwatch-agent#2096 (register k8seventsreceiver)
• feat(container-insights): Add K8s events pipeline to cluster-scraper aws-observability/helm-charts#301 (cluster-scraper events pipeline)

These were the OTEL events tests originally written in an internal scratchpad repo. Porting them into the agent test framework here (alongside the existing PR #671 standard-cluster metrics tests and the in-flight PR #697 logs framework) so they run as part of the standard cluster CI run.

Description of changes

Ports K8s events integration tests into the OTEL standard cluster test suite. Tests validate that K8s events ingested via the k8s_events receiver flow through the cluster-scraper logs pipeline, land in CloudWatch Logs, and carry the expected resource/scope/log attributes set by the pipeline processors (k8sattributes, nodemetadataenricher, transform).

Components:

• util/otellogs/: CW Logs Insights client with query cache and OTLP JSON parser, mirroring the structure of util/otelmetrics. Differs from the in-flight PR feat(otel): Add OTLP logs integration test framework #697 util/otellogs only in that the query cache uses @message like substring filtering — events JSON has dots in the attribute keys (e.g. "k8s.cluster.name" is a literal key) which Insights cannot navigate structurally.
• test/otel/standard/events_test.go: 7 integration tests — log-group existence, resource attributes, cluster identity, scope attributes, log attributes, pod enrichment (count-based majority assertion), and event-kind diversity.
• test/otel/standard/events_setup_test.go: log-group helper, pipeline constant, and logs client init invoked from TestMain.
• terraform/eks/daemon/otel/main.tf: pre-creates the events log group + stream (the OTLP x-aws-log-group-create header is not honored by the backend), enables otelContainerInsights.events.enabled in the helm release, and bumps the test-runner sleep from 3 to 5 minutes for log propagation.

License

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

Tests

Validated end-to-end against a standard EKS cluster with an agent build incorporating the three feature PRs above and the helm-charts #301 events config. Followed the exact Terraform-driven sequence: log group + stream creation → helm install with events enabled → 5-minute propagation wait → go test -tags integration. Result: 7/7 PASS.

=== RUN   TestEventsLogGroupExists                              PASS
=== RUN   TestEventsResourceAttributes                          PASS (10/10 sub)
=== RUN   TestEventsClusterIdentity                             PASS
=== RUN   TestEventsScopeAttributes                             PASS (3/3 sub)
=== RUN   TestEventsLogAttributes                               PASS (6/6 sub)
=== RUN   TestEventsPodEnrichment                               PASS
    receiver_attrs                                              PASS
    enrichment_attrs (k8s.node.name, host.id, host.name,
                      host.type, host.image.id, cloud.az)       PASS (6/6 sub)
    workload_derivation                                         PASS
=== RUN   TestEventsKindDiversity                               PASS
PASS