LTXAVTextEncoderLoader required a diffusion checkpoint to be present even
though the official LTX-2.5 gemma4 text encoder ships as a standalone
file with no matching checkpoint. Add a "none" option so ckpt_name can be
skipped and only the text encoder file is loaded.
LTXVAudioVAELoader read its file list from the checkpoints folder, but
the official Lightricks release ships the audio VAE under models/vae/,
leaving the combo box empty for that layout. Read from the vae folder
instead, matching VAELoader.
Fixes#15595
* Move dataset/text nodes to text category
* Rename category utils into utilities
* Rename category api node into partner
* Move categories conditioning, latent, sampling, model_patches, training, etc. under model category
* Dispatch partner nodes in to 3d, audio, image, text, video categories
* Move PreviewAny node to utilities category
chelsealong
Alexis Rolland
Jukka Seppänen
drozbay
comfyanonymous
Christian Byrne