fix(llmobs): avoid missing ml app error if llmobs is disabled [backport 3.9]

[github-actions]

Backport 17ce936 from #13717 to 3.9.

Our _start_span() method does a bunch of llmobs-instrumentation/modifications under the hood, even if LLMObs is disabled. Specifically our llm/task/tool/agent/retrieval/workflow/embeddings() methods only log a warning if ML_APP is missing, but still route to our _start_span() method, which then continues on to do a ML_APP check (and raises if this is missing), even if LLMObs is disabled. This means if ML_APP is missing, any manual instrumentation will break customer apps if they disable LLMObs.

We need to make _start_span() a no-op wrapper around tracer.trace(...). This fix ensures that if llmobs is disabled, we return the span immediately before we start doing llmobs-specific logic.

Checklist

• PR author has checked that all the criteria below are met
• The PR description includes an overview of the change
• The PR description articulates the motivation for the change
• The change includes tests OR the PR description describes a testing strategy
• The PR description notes risks associated with the change, if any
• Newly-added code is easy to change
• The change follows the library release note guidelines
• The change includes or references documentation updates if necessary
• Backport labels are set (if applicable)

Reviewer Checklist

• Reviewer has checked that all the criteria below are met
• Title is accurate
• All changes are related to the pull request's stated goal
• Avoids breaking API changes
• Testing strategy adequately addresses listed risks
• Newly-added code is easy to change
• Release note makes sense to a user of the library
• If necessary, author has acknowledged and discussed the performance implications of this PR as reported in the benchmarks PR comment
• Backport labels are set in a manner that is consistent with the release branch maintenance policy

[github-actions]

CODEOWNERS have been resolved as:

releasenotes/notes/fix-llmobs-ml-app-disabled-7879e1fcb2d1e0da.yaml     @DataDog/apm-python
ddtrace/llmobs/_llmobs.py                                               @DataDog/ml-observability
tests/llmobs/test_llmobs_service.py                                     @DataDog/ml-observability

[github-actions]

Bootstrap import analysis

Comparison of import times between this PR and base.

Summary

The average import time from this PR is: 278 ± 8 ms.

The average import time from base is: 275 ± 5 ms.

The import time difference between this PR and base is: 2.6 ± 0.3 ms.

Import time breakdown

The following import paths have shrunk:

ddtrace.auto 1.192 ms (0.43%)

ddtrace 0.633 ms (0.23%)

ddtrace.bootstrap.sitecustomize 0.559 ms (0.20%)

ddtrace.bootstrap.preload 0.559 ms (0.20%)

ddtrace.internal.remoteconfig.client 0.559 ms (0.20%)

[pr-commenter]

Benchmarks

Benchmark execution time: 2025-06-23 18:06:06

Comparing candidate commit 2bb4336 in PR branch backport-13717-to-3.9 with baseline commit c90d7fe in branch 3.9.

Found 0 performance improvements and 3 performance regressions! Performance is the same for 550 metrics, 3 unstable metrics.

scenario:iastaspects-lower_aspect

• 🟥 execution_time [+208.567ns; +248.710ns] or [+9.302%; +11.092%]

scenario:iastaspects-rstrip_aspect

• 🟥 execution_time [+1.424µs; +1.519µs] or [+11.136%; +11.885%]

scenario:iastaspects-strip_aspect

• 🟥 execution_time [+1.498µs; +1.589µs] or [+11.543%; +12.242%]