Llm-gpt4all

- Upgrade to latest [gpt4all](https://pypi.org/project/gpt4all/) (2.5.1) which adds support for several new models including...
- `llm -m Meta-Llama-3-8B-Instruct "say hello with a lot of words"` to run the new Llama 3 8B Instruct model! [29](https://github.com/simonw/llm-gpt4all/issues/29)
- Fixed `FileNotFoundError` when running for the first time - thanks, [b421](https://github.com/b421). [#27](https://github.com/simonw/llm-gpt4all/pull/27)

- Now provides access to model options such as `-o max_tokens 3`. Thanks, [Mauve Signweaver](https://github.com/RangerMauve). [#3](https://github.com/simonw/llm-gpt4all/issues/3)
- Models now work without an internet connection. Thanks, [Cameron Yick](https://github.com/hydrosquall). [#10](https://github.com/simonw/llm-gpt4all/issues/10)
- Documentation now includes the location of the model files. Thanks, [Werner Robitza](https://github.com/slhck). [#21](https://github.com/simonw/llm-gpt4all/pull/21)

- Upgrade for compatibility with the latest [gpt4all](https://github.com/nomic-ai/gpt4all). [#16](https://github.com/simonw/llm-gpt4all/issues/16)

The new version of `gpt4all` uses models in the `gguf` format. You may have existing `ggml` models - these will no longer be used by the plugin, so it's safe to delete those files from your `~/.cache/gpt4all` directory.

- Fixed issue preventing this tool from reliably working offline. Thanks, RangerMauve. [9](https://github.com/simonw/llm-gpt4all/pull/9)

- Initial plugin release.