Rename PyPI distribution from mobius-ai to mobius-onnx

The project was renamed, so update the distribution name accordingly. Only
the PyPI/distribution name changes — the import package remains `mobius`
(the `mobius` console script and `import mobius` are unaffected), and the
dynamic version still resolves from `mobius.__version__`.

Update every `pip install mobius-ai[...]` reference across docs, examples,
tests, the CHANGELOG, and the GGUF install hint in `mobius build-gguf`, plus
the `importlib.metadata.version("mobius-ai")` lookup in the model-builder
comparison example.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Signed-off-by: justinchuby <justinchuby@users.noreply.github.com>

Author: justinchuby

CHANGELOG.md

@@ -193,7 +193,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   intentionally skipped tensors (tokenizer, rope_freqs) from
   genuinely unmapped ones.
 - `gguf` optional dependency group in `pyproject.toml`
-  (`pip install mobius-ai[gguf]`).
+  (`pip install mobius-onnx[gguf]`).
 - 28 unit tests for GGUF reader, config mapping, tensor mapping,
   tensor processors, and CLI using synthetic GGUF files.
 

docs/api/build_from_gguf.md

@@ -7,7 +7,7 @@ from mobius import build_from_gguf
 ```
 
 > **Note**: Requires the optional `gguf` package:
-> `pip install mobius-ai[gguf]`
+> `pip install mobius-onnx[gguf]`
 
 ## Signature
 

docs/cli_reference.md

@@ -257,7 +257,7 @@ mobius build --model Qwen/Qwen2.5-0.5B output_dir/ \
 
 Build an ONNX model from a GGUF file (e.g. from llama.cpp).
 
-> **Note**: Requires the optional `gguf` package: `pip install mobius-ai[gguf]`
+> **Note**: Requires the optional `gguf` package: `pip install mobius-onnx[gguf]`
 
 ### Synopsis
 

docs/design/gguf-support-proposal.md

@@ -465,7 +465,7 @@ Add `gguf` as an optional dependency in a new extras group:
 [project.optional-dependencies]
 gguf = ["gguf>=0.10.0"]
 # Combined:
-all = ["mobius-ai[transformers,gguf]"]
+all = ["mobius-onnx[transformers,gguf]"]
 ```
 
 The `gguf` import is lazy — users who don't use GGUF features

docs/getting-started.md

@@ -12,7 +12,7 @@ automatic weight downloading and conversion, including bfloat16 models via
 ## Installation
 
 ```bash
-pip install mobius-ai
+pip install mobius-onnx
 ```
 
 For development and testing:
@@ -112,7 +112,7 @@ mobius build-gguf path/to/model.gguf --output output/model/
 ```
 
 > **Note**: GGUF support requires the optional `gguf` package:
-> `pip install mobius-ai[gguf]`
+> `pip install mobius-onnx[gguf]`
 
 ### Build quantized models (GPTQ/AWQ)
 

examples/diffusion.py

@@ -85,7 +85,7 @@ def main():
     except ImportError:
         print(
             "Error: mobius is not installed.\n"
-            "Install with: pip install mobius-ai[transformers]",
+            "Install with: pip install mobius-onnx[transformers]",
             file=sys.stderr,
         )
         sys.exit(1)

examples/fun_asr.py

@@ -17,7 +17,7 @@
 
 Prerequisites::
 
-    pip install mobius-ai[transformers] sounddevice torchaudio pyyaml
+    pip install mobius-onnx[transformers] sounddevice torchaudio pyyaml
 
 Usage::
 

examples/gemma4_12b_text_ort_genai.py

@@ -40,7 +40,7 @@
 
 Requirements::
 
-    pip install mobius-ai[ort-genai] transformers
+    pip install mobius-onnx[ort-genai] transformers
     # plus a per-layer-KV-aware onnxruntime-genai build
 
 Usage::

examples/gemma4_genai.py

@@ -14,7 +14,7 @@
 
 Requirements::
 
-    pip install mobius-ai[transformers] onnxruntime-genai
+    pip install mobius-onnx[transformers] onnxruntime-genai
 
 Usage::
 

examples/gemma4_multimodal.py

@@ -36,7 +36,7 @@
 
 Prerequisites::
 
-    pip install mobius-ai[transformers]
+    pip install mobius-onnx[transformers]
     pip install torchaudio pillow  # for audio and image loading
 
 Usage::