Fix for various byte compilation issues, with real Vertex issue

The 0.14.0 was released incorrectly, and had a few byte compilation issues, which are now fixed.

Corrected incorrect functions in Vertex and the request library to functions that no longer exist.

Change to way prompts are created

What's Changed

• llm-make-chat-prompt is introduced and obsoletes other ways of creating chat prompts.
• Improved Vertex error handling
• fix: OpenAI API keys passed as multibyte strings by @hraban in #44

New Contributors

• @hraban made their first contribution in #44

Full Changelog: 0.13.0...0.14.0

Claude function calling, provider rewrite, better cleanup and errors

• Add Claude's new support for function calling.
• Refactor of providers to centralize embedding and chat logic.
• Remove connection buffers after use.
• Fixes to provider more specific error messages for most providers.

Warning improvements

• Refactor of warn-non-nonfree methods.
• Add non-free warnings for Gemini and Claude.

Error and streaming fixes

• Send connection issues to error callbacks, and fix an error handling issue in Ollama.
• Fix issue where, in some cases, streaming does not work the first time attempted.

Ollama and Open AI-compatibility fixes

• Fix issue in llm-ollama with not using provider host for sync embeddings.
• Fix issue in llm-openai where were incompatible with some Open AI-compatible backends due to assumptions about inconsequential JSON details.

Claude support

This release adds support for Anthropic's Claude models.

Function calling plus minor fixes

What's Changed

• Added function calling, currently supported just by Open AI and Gemini (and, frankly, Gemini's support has problems)
• Introduce llm-capabilities, which indicates which non-standard functionality each provider supports.
• Fixed issue of logging when we weren't supposed to.
• Change to use Ollama chat endpoint by @tquartus in #16
• Fix emacs 28 incompatibility llm-fake by @s-kostyaev in #21
• Add CI by @s-kostyaev in #24
• Fix ollama mentioned instead of llama.cpp by @SmallAndSoft in #25

New Contributors

• @tquartus made their first contribution in #16
• @s-kostyaev made their first contribution in #21
• @SmallAndSoft made their first contribution in #25

Full Changelog: 0.9.1...0.10.0

Fixes for Gemini, new default for Open AI embedding model

• Default to the new "text-embedding-3-small" model for Open AI. Important: Anyone who has stored embeddings should either regenerate embeddings (recommended) or hard-code the old embedding model ("text-embedding-ada-002").
• Fix response breaking when prompts run afoul of Gemini / Vertex's safety checks.
• Change Gemini streaming to be the correct URL. This doesn't seem to have an effect on behavior.

Token limits

• Add llm-chat-token-limit to find the token limit based on the model.
• Add request timeout customization.