SCHED-292 Tune otel log collector + make it better configurable

[ChessProfessor]

Problem

Large Slurm clusters can generate too many small OpenTelemetry log export requests, increasing load on logging ingest/Envoy. Log public endpoint defaults were also hardcoded to eu-north1, so default log forwarding could ignore observability.region.

Solution

Made OTel log batch timeout, sendBatchSize, and sendBatchMaxSize configurable and raised defaults to 1s / 2000 / 5000. Changed the default public logging endpoint to derive from observability.region, while preserving explicit observability.opentelemetry.publicEndpoint overrides.

Testing

Ran:

- helm template test-release helm/soperator-fluxcd
- helm unittest -f 'tests/opentelemetry_collector_config_test.yaml' helm/soperator-fluxcd
- helm unittest helm/soperator-fluxcd
- helm lint helm/soperator-fluxcd
- git diff --check

Release Notes