Convert dependencies to uv. Add API/MCP server

[snexus]

• Update dependencies and convert the project to uv based package management. As a result, requirements.txt was deleted.
• Remove support for llama-cpp, as the compatible version became obsolete and llamacpp can be used as a standalone package, utilizing OpenAI compatible connection.
• Introduce MCP server for semantic search and RAG answer operations, allowing compatibility with any MCP client

[Copilot]

Pull Request Overview

This PR updates package dependencies to uv‑based management, removes support for llama‑cpp, and introduces an MCP server for semantic search and RAG answer operations.

• Dependency upgrades and configuration updates
• Removal of llama‑cpp support and introduction of MCP server
• Update of configuration and cache handling throughout the codebase

Reviewed Changes

Copilot reviewed 13 out of 19 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
src/llmsearch/webapp.py	Updated import types, configuration variable renaming, and added a workaround for torch classes; potential naming and comment typo issues noted.
src/llmsearch/utils.py	Refactored import statements and minor formatting changes.
src/llmsearch/process.py	Minor formatting and parameter adjustments.
src/llmsearch/models/utils.py	Commented out references to deprecated LlamaModel.
src/llmsearch/config.py	Improved YAML file loading with explicit encoding.
src/llmsearch/api.py	Integrated FastAPI MCP server and updated dependency injections.
sample_templates/*.yaml	Updated configuration values for embeddings and model names.
pyproject.toml	Adjusted Python version requirement and dependency definitions.
README.md	Updated documentation to reflect new features and enhancements.

Files not reviewed (6)

• .flake8: Language not supported
• docker/Dockerfile: Language not supported
• docker/entrypoint.sh: Language not supported
• docs/installation.rst: Language not supported
• docs/usage.rst: Language not supported
• sample_templates/llm/llamacpp.yaml: Language not supported

Comments suppressed due to low confidence (2)

src/llmsearch/webapp.py:88

• The function name 'udpate_index' appears to be misspelled; consider renaming it to 'update_index' for clarity and consistency.

def udpate_index(doc_config_path: str, model_config_file):

src/llmsearch/webapp.py:190

• There is a typo in the comment: 'paratemeters' should be corrected to 'parameters'.

# _config and _bundle are under scored so paratemeters aren't hashed