fix(ci): cap pydantic-ai-slim below 2.34 in unit test requirements - #15701

Checking for broken links...

docs/phoenix/cookbook/datasets-and-experiments/text2sql.mdx
  - images/open-in-colab.svg

docs/phoenix/cookbook/prompt-engineering/prompt-learning-optimizing-prompts-for-classification.mdx
  - images/what-is-prompt-learning.png
  - images/result-accuracy.png

docs/phoenix/datasets-and-experiments/how-to-experiments/eval-ci-with-pytest.mdx
  - datasets-and-experiments/how-to-experiments/run-experiments

docs/phoenix/evaluation/typescript-quickstart.mdx
  - pre-built-metrics/retrieval-relevance

docs/phoenix/integrations/llm-providers/openrouter.mdx
  - images/openrouter-light.png

docs/phoenix/prompt-engineering/how-to-prompts/create-a-prompt.mdx
  - phoenix

docs/phoenix/self-hosting/configuration.mdx
  - self-hosting/features/sandbox-runtimes

Found 8 broken links in 10603.77ms