Add the lightx2v Minimax-h3-Turbo LoRA as a per-request alternative

Switch to the recommended ckpt850 EMA LoRA; default steps back to 4

Fix LoRA toggle (zip misuse made converted a tuple), add a queue/ETA progress bar

Default to the ckpt500 turbo weights, add a per-request LoRA toggle

Fold the 4-step Turbo LoRA into the transformer, default to 4 steps