Pass metrics observations to upstream SDKStats Manager

[rads-1996]

No description provided.

[github-actions]

Performance comparison

Threshold: regressions >15.0% on gating scenarios fail the build. Higher ops/s is better; positive Δ means the PR is slower.

Scenario	Gating	Baseline (ops/s)	Candidate (ops/s)	Δ %	Status
azure_monitor_log	yes	21,468.9	21,702.8	-1.08%	✅
azure_monitor_span	yes	159,744.4	163,292.0	-2.17%	✅
otel_log	no	25,512.2	25,469.3	+0.17%	✅
otel_span	no	47,494.7	46,996.9	+1.06%	✅

[Copilot]

Pull request overview

This PR updates the distro’s SDKStats “network” metric integration so that OTLP/Agent365 request-success counts are contributed to the upstream Azure Monitor Statsbeat metric stream via StatsbeatManager.add_metric_callback, rather than by mutating upstream observable gauge callback lists directly.

Changes:

• Refactors network SDKStats integration to register an additional observation callback through StatsbeatManager.add_metric_callback.
• Updates module documentation/wording to reflect “observations” rather than “gauges”.
• Removes the prior thread-safe idempotency guard and test reset helper (impacting existing test expectations).

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.