chore: force clean Hugging Face Space rebuild

Fix past-report bleed on no-mappable sections with notes-bounded baseline.

fix: shorten HF Space short_description to meet 60-char limit

Deploy v2 backend to Hugging Face Spaces.

chore: ignore v2 runtime data and local test artefacts

Add demo documentation and Docker setup for v2 report generation system

Fix auth on notes upload and DOCX export; stabilize HF token signing.

Ship production RAG hardening: citation extraction, full-library retrieval, auth.

Disable LLM RAG sanitisation by default on HF Spaces to fix slow ingestion.

Speed up generation and ingestion; add live report progress on /status.

Default report generation to standard RAG/LLM on e561e67 baseline with /health verification.

feat: RAG upload sanitisation layer before vector indexing

chore: add HB-BS standard paras reference doc (Git LFS)

chore: add pytest output log

feat: async pipeline, job queue, generation hardening, and docs

feat: AI interference levels (minimum|medium|maximum) UI, API, prompts, generation

feat: enhance embedding model configuration and retrieval logic

feat: introduce strict uploaded only mode and enhance logging for AI generation

feat: enhance AI involvement and standard paragraph handling

feat: enhance AI involvement handling and transparency metrics

fix: enhance section text update handling and validation

fix: apply structured section blocks to L1/L2 as well

fix: restore L3 structured section blocks and label survey level

fix: allow reference-only RAG guidance in notes-only mode

fix: notes-only generation to prevent RAG content leak

deploy: harden inspector + add RAGAS eval harness

refactor: enhance non-invention guard and tier-aware output handling

fix: stop false Level 1 classification of messy Level 3 field notes

fix: add pypdf and openai as explicit production dependencies

fix: use same-origin API base so uploads work on HF Spaces

fix: add python-docx to production dependencies

chore: add HF Spaces YAML frontmatter

chore: update Docker configuration and application settings for improved deployment

chore: update .gitignore and README for improved clarity and functionality

feat: adaptive writing style — user docs first, KB fallback, few-shot examples

feat: implement two-layer hallucination grounding check with 21 new tests

- Added support for RICS survey levels (1, 2, 3) in document uploads and reports, allowing for better tier management and retrieval filtering.

Added canonical rollout functionality, including new API endpoints and request/response models for upgrading indexed content. Updated main application to include the canonical rollout router. Introduced a new document loading function for raw documents in the ingestion pipeline. This enhances content management and semantic upgrades across the application.

- Introduced a new API endpoint for finding similar content in the tenant's workspace, allowing users to identify overlapping uploads and draft notes.

- Replaced timezone.utc with UTC from datetime for consistent datetime handling.

feat: FAISS-only vector store (remove Chroma)

- Introduced `ingest_timeout_seconds` setting in config.py to define maximum time for document ingestion.

Merge origin/experiment-lab (LangChain RAG) into experiment-lab + product features

feat(experiment-lab): AI transparency bar, tenant-scoped RAG, citations, proofread/enhance fixes

- Updated .gitignore to exclude proprietary RICS documents.

- Updated `.env.example` to include new settings for bulk upload limits and batch processing.

feat: enhance document upload and processing capabilities with new personalisation features

feat(experiment-lab): rewrite RAG pipeline using LangChain framework

fix: comprehensive codebase review fixes (critical + major + minor issues)

fix: apply all bug fixes and improvements (embeddings, cache, postprocess, frontend, tests)