[DRAFT] Initial /v1/messages implementation

[RobGeada]

Description

Adds /v1/messages support to NeMo Guardrails, namely:

• A native /v1/messages client, mirroring the native (non-langchain) OpenAI client, which allows for native communication with /v1/messages servers for inference, self-check, etc
• A /v1/messages endpoint in the server, allowing for proxying of /v1/messages calls
• A /v1/messages/checks endpoint: /v1/chat/completions is to v1/guardrail/checks as /v1/messages is to /v1/messages/checks

As well as unit tests verifying:

• all possible fields in the messages are sent transparently through NeMo Guardrails
• all fields in the pinned anthropic package are included in the API's AnthropicMessageRequest BaseModel

Related Issue(s)

#2226

Verification

All unit tests pass

Have manually verified:

• Proxying to a /v1/messages model works as expected
• /v1/messages models can be used as self-check models
• Content is correctly extracted from a /v1/messages request and sent to guardrails
• /v1/messages/checks works as expected/
• Have successfully guardrailed Claude Code without noticing any obvious issue

AI Assistance

• No AI tools were used.
• AI tools were used; a human reviewed and can explain every change (tool: Claude Code).

Checklist

• I've read the CONTRIBUTING guidelines.
• This PR links to a triaged issue assigned to me.
• My PR title follows the project commit convention.
• I've updated the documentation if applicable.
• I've added tests if applicable.
• I've noted any verification beyond CI and any checks I couldn't run.
• I did not update generated changelog files manually.
• I addressed all CodeRabbit, Greptile, and other review comments, or replied with why no change is needed.
• @mentions of the person or team responsible for reviewing proposed changes.