feat(llmobs): activate APM scope for standalone agent spans

[yahya-mouman]

Summary

• When an LLMObs agent span is its own APM local root (no ambient APM parent), it now activates its underlying APM span as the current scope via standaloneApmScope.
• Without this, each new DDLLMObsSpan inside the same workflow starts a fresh APM root with a different trace ID, causing the trace-ID consistency gate in the DDLLMObsSpan constructor to reject parent_id and session_id inheritance — resulting in a flat LLMObs trace with no hierarchy.
• When an agent span already has an ambient APM parent (non-standalone), standaloneApmScope is null and no additional scope is opened.

This is a follow-up to #12238 (agent attribution), extracted into its own PR because it's a standalone behavioral change with its own tests.

Test plan

• New JUnit tests in DDLLMObsSpanStandaloneApmScopeTest:
  • standaloneAgentSpanActivatesApmScopeForOutgoingPropagation — verifies standaloneApmScope is non-null when no ambient APM parent
  • nonStandaloneAgentSpanDoesNotActivateApmScope — verifies standaloneApmScope is null when an APM parent already exists
• Groovy test cleanup: added cleanup: test.finish() to the two DDLLMObsSpanTest tests that use LLMOBS_AGENT_SPAN_KIND to prevent scope leakage

🤖 Generated with Claude Code

[datadog-datadog-prod-us1-2]

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 58.78% (-0.06%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 1207e73 | Docs | View more details | Give us feedback!}

[dd-octo-sts]

🟢 Java Benchmark SLOs — All performance SLOs passed

Suite	Status
Startup	🟢 pass

SLO thresholds are defined here based on automatically generated metrics. A warning is raised when results are within 5% of the threshold.

PR vs. master results

Scenario	Candidate	master	Δ (95% CI of mean)
startup:insecure-bank:iast:Agent	13.99 s	14.00 s	[-0.9%; +0.8%] (no difference)
startup:insecure-bank:tracing:Agent	12.89 s	13.00 s	[-1.6%; -0.1%] (maybe better)
startup:petclinic:appsec:Agent	17.62 s	17.17 s	[+1.7%; +3.6%] (significantly worse)
startup:petclinic:iast:Agent	17.53 s	17.63 s	[-1.7%; +0.5%] (no difference)
startup:petclinic:profiling:Agent	17.45 s	17.43 s	[-1.0%; +1.2%] (no difference)
startup:petclinic:sca:Agent	17.37 s	17.41 s	[-1.4%; +0.9%] (no difference)
startup:petclinic:tracing:Agent	16.66 s	16.59 s	[-0.5%; +1.4%] (no difference)

Commit: 1207e73c · CI Pipeline · Benchmarking Platform UI

---

Load and DaCapo benchmarks can be triggered manually in the GitLab pipeline. Results will appear in the Benchmarking Platform UI after completion.