Fix MPS tensor dtype conversion in audio decoding

[Copilot]

Audio decoding fails on Apple Silicon with invalid type: 'torch.mps.FloatTensor' because PyTorch implicitly converts token indices from torch.long to torch.float when moving tensors to MPS devices.

Changes

• backend/app/services/music_service.py:850: Explicitly preserve torch.long dtype when moving frames to codec device

# Before: implicit dtype conversion on MPS
frames_for_codec = frames.to(device=pipeline.codec_device)

# After: explicit dtype preservation
frames_for_codec = frames.to(device=pipeline.codec_device, dtype=torch.long)

The frames tensor contains token IDs used as indices in codec.detokenize() and must remain integer type. Pattern matches existing MPS handling at lines 1585, 1603.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Decoding Audio Failure</issue_title>
<issue_description>After PR #14 generation succeeds at a reasonable speed, but once completed it fails with:

Deconding Audio:
Generation Failed: invalid type: 'torch.mps.FloatTensor'

Lets fix this last bug to make sure our generations are correctly decoded to audio.</issue_description>

<agent_instructions>Generation still fails in the last conversion phase:

Generation Failed: invalid type: 'torch.mps.FloatTensor'

Resolve this issue and make sure audio conversion succeeds 100%</agent_instructions>

Comments on the Issue (you are @copilot in this section)

@lmangani This change was ineffective and caused a new issue:

Generation Failed: tensors used as indices must be long, int, byte or bool tensors</comment_new>
<comment_new>@lmangani
Generation still fails in the last conversion phase:

Generation Failed: invalid type: 'torch.mps.FloatTensor'

• Fixes Decoding Audio Failure #15

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.