Add year-over-year trends + expanded ratios (EBITDA, turnover, FCF conversion)

Add deterministic computed-ratios layer (Approach 2); FinanceBench 20% -> 45%

Add hybrid RAG: XBRL structured-financials layer for numeric questions

Expand corpus to 25 companies (add Apple, Alphabet, Amazon, NVIDIA, Tesla, JPMorgan, Walmart)

Expand corpus to 18 companies (recent SEC 10-Ks); ship prebuilt index via git-lfs

Pin Python 3.12 + verified package versions; store index in /tmp on Linux

Fix cold-start crash: probe vector store via filesystem, not a Chroma client

FinChat: RAG chatbot over SEC 10-K filings

Duplicate from streamlit/streamlit-template-space