feat(llamacpp): Completion Reasoning support

[njbrake]

Description

Works with our existing logic!

To set up llama cpp I did

 brew install llama.cpp
llama-server -hf ggml-org/Qwen3-1.7B-GGUF --jinja --port 8090

Make sure you don't have LLAMA_API_KEY set, otherwise it'll add auth to the llama cpp server which we don't want.
That's an annoying thing, the Meta Llama API and Llama.cpp share the same API key name.

PR Type

🆕 New Feature ## Relevant issues

Checklist

• I have added unit tests that prove my fix/feature works
• New and existing tests pass locally
• Documentation was updated where necessary
• I have read and followed the contribution guidelines```

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Files with missing lines	Coverage Δ
src/any_llm/providers/llamacpp/llamacpp.py	100.00% <100.00%> (ø)

... and 34 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.