refactor(agent): enable immediate response instruction in supervisor prompt

refactor(api): switch LLM provider from MiniMax to Google Generative AI

refactor(api): add LLM abstraction layer to replace hardcoded model strings

feat(tools): add http_get tool for web content fetching

feat(tools): add ocr_reader and list_files tools to expand agent capabilities

feat(agent): add file URL support to agent query handling

refactor(agent): remove unused websearch_agent and update supervisor instructions

feat(run): add retry mechanism with error handling to agent entry point

feat(agent): add answer extraction module and integrate into run function

refactor(agent): split websearch agent into separate modules

refactor(agent): prioritize accuracy in math solver tool instructions

chore(agent): enhance supervisor prompts and increase websearch recursion limit

feat(agent): implement supervisor agent with tool routing and math solver

feat(agent): add web search agent with LangGraph and Tavily integration