Fix CPU training + 1902ms step speed bottleneck

fix: remove torch from requirements to preserve pre-installed CUDA torch

fix: install CUDA-enabled torch for GPU spaces

pre-deployment fixes for HF Spaces

Fix gradio/huggingface_hub version conflict on Python 3.13

Add audioop-lts for Python 3.13 pydub fix (cache bust)

Upgrade to gradio 5 for Python 3.13 compatibility

Fix Python 3.13 audioop: use audioop-lts

Fix Python 3.13 audioop compatibility

SpindleFlow RL — periodic push + log persistence