BUG: Cache validation does not check latent size for SD/SDXL (sd3 branch)

[feffy380]

The new cache validation in the sd3 branch only checks the key name. When multiresolution is disabled (like with sd and sdxl training), the key is latents and the size of the tensor is not checked. If the training resolution or bucketing changes, the cache should not pass validation.

sd-scripts/library/strategy_base.py

Lines 446 to 451 in 123474d

	key_reso_suffix = f"_{expected_latents_size[0]}x{expected_latents_size[1]}" if multi_resolution else ""
	try:
	npz = np.load(npz_path)
	if "latents" + key_reso_suffix not in npz:
	return False