feat(scores): support text scores in v3

[wochinge]

Summary

• Backports TEXT score support to the v3 SDK score creation APIs and public score typing.
• Adds v3 generated API-style TEXT score enums, models, response unions, and exports so score and score-config responses can deserialize TEXT values.
• Keeps experiment evaluator score typing limited to numeric, categorical, and boolean scores.
• Adds live coverage for creating a TEXT score and creating/listing a TEXT score config.
• Backport from: feat(api): update API spec from langfuse/langfuse 1999706 #1615

Linear

• LFE-9862

Major Decisions

• Applied the TEXT score API changes as a targeted v3 generated-style patch instead of regenerating from the current Fern spec, because the v3 generated API layout has drifted from current main and a full regeneration would introduce unrelated churn.

Review Focus

• TEXT variants in the generated API-style unions and exports under langfuse/api/resources.
• SDK overload/type behavior for string score values, especially preserving categorical defaults while allowing explicit TEXT.
• ExperimentScoreType usage to avoid unintentionally enabling TEXT scores in experiment evaluator typing.
• Live score-config test behavior, including archiving the created TEXT config after assertions.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: adec90aa51

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".