### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

### Actual behavior

When OTel v2 is configured with `mapper_names: [langfuse]`, `LangfuseMapper` emits generation model/usage/cost plus team metadata, but it does not map the standard Langfuse trace controls already supported by LiteLLM's existing Langfuse OTel integration:

- `trace_user_id` / end-user identity → `user.id` or `langfuse.user.id` - `session_id` → `session.id` or `langfuse.session.id` - `trace_name` → `langfuse.trace.name` - request tags → `langfuse.trace.tags` - explicitly approved trace metadata → `langfuse.trace.metadata.<key>`

As a result, native OTel v2 traces arrive in Langfuse with correct tokens and cost but without top-level user/session attribution, custom trace names, or tags. The same request fields are supported by `LangfuseOtelLogger._set_metadata_attributes` in the existing integration, so switching to the v2 mapper loses semantics.

### Expected behavior

The v2 Langfuse mapper should preserve the established LiteLLM Langfuse metadata contract using the current official Langfuse OTLP attribute names. Trace-wide …