Deploy qmd-web

Deploy qmd-web

Deploy qmd-web

Deploy qmd-web

feat: align browser demo with qmd pipeline

fix: expansion test signatures and tokenizer type cast for tsc

feat: manual generation for no-KV-cache model, fix blend scoring, add coffee doc

feat: use fine-tuned 1.7B expansion model with chunked external data

feat: switch expansion model to Qwen3-0.6B-ONNX

fix: add timeout for expansion model loading, rebuild

fix: use self-contained ONNX models, sequential loading, Vite config

fix: model loading for browser compatibility

add pipeline orchestrator as AsyncGenerator yielding stage events

feat: add RRF fusion and position-aware score blending

add reranking module using Qwen3-Reranker yes/no token scoring

add query expansion module with lex/vec/hyde parsing

add embeddings and vector search modules

fix: use AutoModelForCausalLM for reranker (yes/no token scoring)

add model loading layer for Transformers.js WebGPU pipelines

add in-memory BM25 search index over chunks

add chunking module with markdown-aware token-based splitting