Fix Apr 05 integration test regressions (cuda sm_86) (#48170)

* [MusicgenMelody] Add ("cuda", 8) expected values for 8 integration tests (cuda sm_86)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* [StableLm] Add ("cuda", 8) expected values for 2 integration tests (cuda sm_86)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* [Whisper] Add ("cuda", 8) expected transcription for test_speculative_decoding_non_distil (cuda sm_86)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* [EomtDinov3] Update stale ("cuda", 8) expected values in test_inference_bf16 (cuda sm_86)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* [GLM] Update stale ("cuda", 8) expected text in test_model_9b_eager (cuda sm_86)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* [Dia] Add ("cuda", 8) expected token tensors in test_dia_model_integration_generate_audio_context (cuda sm_86)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* [RAG] Add ("cuda", 8) expected outputs for token/sequence beam search integration tests (cuda sm_86)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* Revert dia, eomt_dinov3, musicgen_melody, rag, whisper — unreliable Apr 04 CI report

These models had success=0 or were missing from the April 04 CI report,
making the April 05 regression classification unreliable. Whisper was
also already fixed on main.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>