fix: overhaul follow-up detection for conversation memory + add diagnostic logs

fix: switch HF endpoint to universal router URL (auto-selects provider)

fix: add stream error handling + startup diagnostics, bump CACHEBUST

fix: bump CACHEBUST to force Docker rebuild with updated model config

feat: overhaul model routing — GLM-5.1 primary, strict 4-model fallback chain, purge all stale refs

fix: update deprecated Hugging Face api-inference URL to router.huggingface.co

fix: update Qwen model identifier to Qwen3.5-9B due to deprecated support

fix: pass HF_API_KEY as HF_TOKEN to avoid Hub rate limits when loading models

feat: Add full multi-turn conversation memory and context rewriting

style(ui): align config dropdowns equidistantly and add padding before arrows

feat(ui): implement custom premium glassmorphic dropdowns for configuration

style(info): use cleaner metrics and bulleted domain explanations in info modal

docs(info): update operational capacity stats and data categories in info modal

feat(ui): complete UI overhaul with smooth animations, glowing elements, and refined citation rendering

feat(ui): restore info modal overlay and limit category select to user-defined options

feat(ui): add full CS category options and publication year filter to settings

feat(ui): add layout collapsibility, sticky status, github footer, and scroll-to-bottom controls

fix(ui): enforce block math newlines for centered render and tweak typing animation

fix(ui): improve citation extraction to match floating unbracketed arxiv IDs

fix(ui): fix citation links rendering in markdown and bypass katex parsing errs

feat(ui): add markdown parsing with math, typewriter text streaming, scrollbar shift, and history tracking actions

fix(hf): add HuggingFace Space README config with docker sdk

fix(backend): bust docker cache + fix async streaming generator to not block event loop

feat(ui): add polished glassy design, framer motion mechanics, and api status header

fix(ui): remove undefined yearFilter from frontend api call

fix(api): improve stream token chunks robustness and sync year filter

feat: implement ChatGPT style UI, streaming generation, math/code blocks, citations & multi-model fallback

fix: update qdrant-client and pydantic to match db schema versions

fix: f-string syntax error in reranker block for python 3.10 compatibility

fix: clone dataset before creating subdirectories

fix: use git clone for dataset instead of hf-cli to bypass rate limits

fix: use minimal requirements to fix Docker build

feat: download 4.4GB data from dataset on build

chore: update project dependencies in requirements.txt

chore: update project dependencies in requirements.txt

feat: add Dockerfile and requirements.txt for HuggingFace deployment

feat: ResearchPilot v2 — Next.js frontend, 358k embeddings pipeline, LaTeX math rendering, dual-GPU Kaggle scaling

feat: FastAPI backend complete

feat: RAG pipeline complete - ResearchPilot end-to-end working

feat: retrieval optimization pipeline complete

feat: vector database indexing complete

fix: exclude data directories from version control

feat: document chunking pipeline complete

feat: data ingestion and processing pipeline complete