- Fix missing indentation in async with semaphore block
- Fix invalid elif syntax in load_mode determination
- Fix request.steps reference (field doesn't exist in request model)

- Without --loadall: serialize all requests (one at a time)
- With --loadall: allow one concurrent request per model

These parameters are not available in the installed version of
stable-diffusion-cpp-python. The engine will auto-detect the
architecture from file headers.

- Added is_huggingface_model_id() helper to detect HF model IDs
- Added download_huggingface_model() to download from HF Hub
- Updated download logic to handle model IDs, URLs, and local paths

- Renamed CLI argument from --clip-l-path to --llm-path
- Updated all references from args.clip_l_path to args.llm_path
- Changed config dictionary key from 'clip_l_path' to 'llm_path'
- Added model_type='z-image' and backend='vulkan' parameters

The bug was that the code checked 'if clip_l_path:' but clip_l_path
was just set to None above, so the download block was always skipped.

Changed to check 'if args.clip_l_path:' to properly detect when the
--clip-l-path CLI argument is provided and trigger the download/caching.

- Changed incorrect 'elif clip_l_path:' to proper 'if clip_l_path:' logic
- Fixed vae_path assignment to use args.vae_path instead of undefined vae_path variable
- This ensures CLIP LLM and VAE paths are properly passed when using
  --clip-l-path and --vae-path CLI arguments with Vulkan backend

These are txt2img() parameters, not constructor parameters

When a URL is passed to --clip-l-path or --vae-path, the model is now automatically downloaded and cached.

- Added --clip-l-path for specifying CLIP LLM model path
- Added --vae-path for specifying VAE model path
- Added --image-sample-method (default: res_multistep for Z-Image Turbo)
- Added --image-steps (default: 4 for Z-Image Turbo)
- Added --image-width (default: 512)
- Added --image-height (default: 512)
- Added --image-cfg-scale (default: 1.0 for Z-Image Turbo)

- When --debug is enabled, show full command line coderai was called with
- Fixed GGUF image model key to use cached file path instead of URL
  (lines 4565 and 5124 now use model_path)
- Removed redundant model_key assignment before model_path resolution

When GGUF image model fails to load with llama.cpp, try loading with stable-diffusion-cpp-python (sd.cpp) as fallback.

Tell user to update llama.cpp instead of falling back to diffusers.

This reverts commit 596d1245.

If llama.cpp fails to load a GGUF image model (e.g., unsupported architecture like lumina2), try loading via diffusers instead.

GGUF files can have different version bytes after 'GGUF' (e.g., GGUF\x03 for version 3).
Changed magic byte check from exact match to prefix check.

Changed condition from 'only if no audio models' to 'always when image_models is configured'.
This ensures the image model downloads at startup even when audio models are present.

- Removed redundant 'import os' statements inside functions (lines 4522, 4926, 5005)
- Added back missing 'from llama_cpp import Llama' that was accidentally deleted
- Global 'import os' at line 12 is now the only one

This fixes the UnboundLocalError when running --list-cached-models or other CLI options.

- --list-cached-models: List all cached models with sizes
- --remove-all-models: Remove all cached models
- --remove-model <modelid>: Remove specific model by name/hash (partial match)

- Check if downloaded file is valid GGUF (magic bytes = 'GGUF')
- If not valid, show clear error that URL is wrong (returns HTML instead)
- Explain that URL must be direct download link ending in .gguf

- Enable verbose=True in llama.cpp to see actual error
- Print GGUF model file size for debugging
- Add try/except with traceback to see detailed errors