Fix text encoder payload parsing and add remote encoder switch

Fix HF auth propagation for text encoder loads

Fix: explicit HF_TOKEN passing to text encoder subprocess

Switch LLM2Vec to Meta-Llama-3.1-8B-Instruct (accessible with current token)

Use mirrored LLM2Vec base model repo and env overrides

Space: keep existing embedding pipeline and start local text-encoder API server on boot

Space: invoke @spaces.GPU function during startup to satisfy ZeroGPU policy

Space: restore stable native-direct startup; keep explicit Qwen 09 controls

Space: redirect to native proxy without iframe recursion; keep Qwen 09 direct load

Space: restore minimal @spaces.GPU hook for ZeroGPU with native-direct UI

Space: run native Kimodo UI directly on PORT to avoid iframe recursion

Space: auto-retry native proxy load and add direct Qwen example button

Space: force native UI proxy and remove fallback iframe switching

Default HF runtime to CPU and honor KIMODO_DEVICE

Force TEXT_ENCODER_MODE=api on Space to avoid gated local fallback

Log imported kimodo module path in Space boot

Boot native demo via proxy and include Qwen example UI flow

Simplify Space app to native Kimodo UI host

Qwen loop: colored timeline blue→red→yellow→green + stop button + spinner

Add spaces GPU-decorated execution handler

Use self-contained UI with Kimodo 3D viewer tab

Deploy Kimodo hackathon UI entrypoint

feat: deploy Card 10 frontend shell to HF Space

feat: lightweight non-LFS Movimento app deployment (#2)

initial commit