feat: Fireworks as inference option

[mislavivanda]

Summary by cubic

Adds Fireworks as an inference option across Python and TypeScript examples, including the LangGraph plan-and-execute agent. Users can choose models via a provider-agnostic MODEL env and run agents with the matching API key.

• New Features
  • Python (LangChain + LangGraph data agents): Model set via MODEL=provider:model (default Anthropic). Switched to init_chat_model. Added Fireworks setup (langchain-fireworks), FIREWORKS_API_KEY, and example fireworks:accounts/fireworks/models/kimi-k2p6.
  • Python (ai-data-analyst): Added FIREWORKS_AI_API_KEY and Fireworks model option in README.
  • TypeScript (mastra/coding-agent): Added Fireworks models to supported list and FIREWORKS_API_KEY to .env.example.
  • TypeScript (vercel-ai-sdk/multi-language-benchmark-agent): Added provider switching guide with @ai-sdk/fireworks@2, example accounts/fireworks/models/kimi-k2p6, and env key setup.

^{Written for commit 2b6d7a9. Summary will update on new commits.}

[cubic-dev-ai]

No issues found across 8 files

_{Re-trigger cubic}