chore(llmobs): add base datasets for experiments

[Kyle-Verhoog]

Adds the basic support for creating and pulling dataset objects from the new Datadog LLMObs Experiments product.

Testing is done using the new cassette logic added in 1.27.4 to the test agent which enables us to easily record requests made to the Datadog API. Instructions for running the tests are included at the top of the tests file.

Dependent on DataDog/dd-apm-test-agent#223 being merged + released.

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:

ddtrace/llmobs/_experiment.py                                           @DataDog/ml-observability
tests/llmobs/llmobs_cassettes/datadog/datadog_api_unstable_llm-obs_v1_datasets_delete_post_0e577f3f.yaml  @DataDog/ml-observability
tests/llmobs/llmobs_cassettes/datadog/datadog_api_unstable_llm-obs_v1_datasets_delete_post_15fc0d15.yaml  @DataDog/ml-observability
tests/llmobs/llmobs_cassettes/datadog/datadog_api_unstable_llm-obs_v1_datasets_e3608c37-7422-4a91-8897-9dc6097e86b4_records_get_ce4912d9.yaml  @DataDog/ml-observability
tests/llmobs/llmobs_cassettes/datadog/datadog_api_unstable_llm-obs_v1_datasets_filter_name__test-dataset-non-existent_get_ceb7b7a1.yaml  @DataDog/ml-observability
tests/llmobs/llmobs_cassettes/datadog/datadog_api_unstable_llm-obs_v1_datasets_filter_name__test-dataset_get_3012b2d9.yaml  @DataDog/ml-observability
tests/llmobs/llmobs_cassettes/datadog/datadog_api_unstable_llm-obs_v1_datasets_post_3335cc3f.yaml  @DataDog/ml-observability
tests/llmobs/llmobs_cassettes/datadog/datadog_api_unstable_llm-obs_v1_datasets_post_816b83e0.yaml  @DataDog/ml-observability
tests/llmobs/test_experiments.py                                        @DataDog/ml-observability
.gitlab/services.yml                                                    @DataDog/python-guild @DataDog/apm-core-python
ddtrace/llmobs/__init__.py                                              @DataDog/ml-observability
ddtrace/llmobs/_constants.py                                            @DataDog/ml-observability
ddtrace/llmobs/_llmobs.py                                               @DataDog/ml-observability
ddtrace/llmobs/_writer.py                                               @DataDog/ml-observability
docker-compose.yml                                                      @DataDog/apm-core-python
tests/llmobs/_utils.py                                                  @DataDog/ml-observability
tests/llmobs/conftest.py                                                @DataDog/ml-observability
tests/llmobs/suitespec.yml                                              @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: 285 ± 4 ms.

The average import time from base is: 290 ± 7 ms.

The import time difference between this PR and base is: -5.0 ± 0.3 ms.

Import time breakdown

The following import paths have shrunk:

ddtrace.auto 2.246 ms (0.79%)

ddtrace.bootstrap.sitecustomize 1.554 ms (0.55%)

ddtrace.bootstrap.preload 1.554 ms (0.55%)

ddtrace.internal.remoteconfig.client 0.716 ms (0.25%)

ddtrace 0.692 ms (0.24%)

ddtrace.internal._unpatched 0.035 ms (0.01%)

json 0.035 ms (0.01%)

json.decoder 0.035 ms (0.01%)

re 0.035 ms (0.01%)

enum 0.035 ms (0.01%)

types 0.035 ms (0.01%)

[pr-commenter]

Benchmarks

Benchmark execution time: 2025-07-09 18:03:19

Comparing candidate commit 06c2352 in PR branch llmobs/dne with baseline commit 6eb491c in branch main.

Found 0 performance improvements and 1 performance regressions! Performance is the same for 547 metrics, 2 unstable metrics.

scenario:telemetryaddmetric-1-distribution-metric-1-times

• 🟥 execution_time [+401.419ns; +431.735ns] or [+13.846%; +14.892%]

[Yun-Kim]

Some initial thoughts!

[brettlangdon]

changes to .gitlab/services.yml and docker-compose.yml lgtm

This change will depend on fixes in #13911 in order to be merged.

[gary-huang]

lgtm for the first pass