### Check for existing issues

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

### What happened?

When calling a vLLM passthrough route (e.g. `POST /vllm/pooling`) where the `model` in the request body resolves to a deployment in `model_list`, the request succeeds (200 OK) but is **never recorded in `LiteLLM_SpendLogs`** (and therefore never shows in the UI Logs tab). Every call logs two non-blocking errors: `_PROXY_track_cost_callback` fails with `standard_logging_object not found`, and the Prometheus callback fails with `standard_logging_object is required, got=None`.

The docs state that Logging is supported for vLLM passthrough routes (https://docs.litellm.ai/docs/pass_through/vllm), but with router models nothing is logged at all.

Note this is related to, but distinct from, #24204 / PR #24205: there, `allm_passthrough_route` rows ARE written with `spend: 0.0`. Here the `StandardLoggingPayload` is never built, so the SpendLogs write is aborted entirely and the request is invisible to spend tracking, budgets, and the UI.

### What I found

`vllm_proxy_route` detects the body model as a router model (`is_passthrough_request_using_router…