Use fine-tuned model v0.1 (WebLLM, constrained JSON decoding, q0f16 on desktop / q4 on phones); MLC converter; v0.3 pipeline

MODEL_LLM.md: v0.3 run needs --max-len 2560 (measured)

MODEL_LLM.md: datasets and their versions

MODEL_LLM.md: fine-tuning versions, current run and next steps

Link the model role document

Add docs/MODEL_LLM.md: role of the language model and how to improve it

v0.3: phone check (iPhone, Android) as the Space front page, catalog 0.3 (74 checks), shared vulnerability rules, real-CVE training data

Open source: Apache-2.0 code, CC BY 4.0 data, NOTICE and CITATION.cff

Offline vulnerability matching: inventory in collectors, signed OSV/NVD/KEV/EPSS bundle builder, matcher in the app, tests

Collectors 0.2: user account type, USB device access, pending updates, outdated apps, software inventory; controls document

Catalog, library and synthetic data v0.2: user account type, USB device access, vulnerabilities

Desktop app (Electron + WebLLM), collector progress output, offline LoRA training script

Steps 2-3: control library, training data builder and faithfulness check

Remove unused template stylesheet

Stage 1 step 1: read-only collectors, evidence schema, catalog and synthetic data

initial commit