Implement Semantic Consolidator V0

- add deterministic canonicalization support for extraction items
- add facts-only semantic consolidation using local Ollama
- preserve source evidence and validate complete fact coverage
- add conservative merge rules and non-LLM tests
- record the first validated real-life consolidation benchmark
- document current scope, limitations and next evaluation step

PROJECT_KNOWLEDGE.md

@@ -17,6 +17,11 @@ Implemented:
 - Transcript normalization in `src/meeting_lab/normalization/`.
 - Technical chunking in `src/meeting_lab/chunking/`.
 - Local per-chunk extraction in `src/meeting_lab/extraction/extract_chunks.py`.
+- Deterministic Canonicalizer V1 in
+  `src/meeting_lab/consolidation/canonicalize.py`.
+- Semantic Consolidator V0 in
+  `src/meeting_lab/consolidation/consolidate_facts.py` for facts-only
+  semantic duplicate detection.
 - Prompt loading from `src/meeting_lab/llm/prompts.py`.
 - Interim Markdown protocol generation in `src/meeting_lab/protocol/`.
 - Non-LLM unit tests for chunking, extraction helpers, protocol rendering and
@@ -30,8 +35,8 @@ Experimental/prototype:
 
 Planned:
 
-- Deterministic Canonicalizer for extraction results.
-- Semantic Consolidator for evidence-preserving semantic merging.
+- Broader semantic consolidation for topic grouping, contradiction handling,
+  uncertainty marking and durable/transient separation.
 - Canonical Meeting Knowledge implementation as the semantic source of truth.
 - Final Working Protocol / Arbeitsprotokoll, Distribution Protocol /
   Verteilerprotokoll and Knowledge Objects / Wissensdatenbankeintrag renderers.
@@ -123,17 +128,37 @@ Chunk Extractions
   -> Output View Renderers
 ```
 
-The Deterministic Canonicalizer is planned Python code with no LLM. It should
-validate and normalize extraction objects, assign stable source references and
-IDs, normalize category names and basic field structure, perform only safe
-deterministic cleanup, optionally group exact duplicates, and preserve all
+Canonicalizer V1 is implemented Python code with no LLM. It validates and
+normalizes extraction objects, assigns stable source references and IDs,
+normalizes category names and basic field structure, performs only safe
+deterministic cleanup, optionally groups exact duplicates, and preserves all
 source evidence. It must not perform uncertain semantic merging.
 
-The Semantic Consolidator is planned local-LLM work. It should merge
-semantically equivalent statements, group content by topic, preserve evidence
-from all contributing chunks, mark contradictions and uncertainty, separate
-durable information from transient discussion, and produce Canonical Meeting
-Knowledge. It does not directly write a protocol.
+CLI:
+
+```text
+PYTHONPATH=src .venv/bin/python -m meeting_lab.consolidation.canonicalize \
+  samples/whisper/meeting_speech_cleaned_chunks \
+  -o /tmp/canonicalized_extractions.json
+```
+
+Semantic Consolidator V0 is implemented as narrow local-LLM work for fact
+items only. It merges semantically equivalent fact statements, preserves source
+references and evidence, and prefers false negatives over false-positive
+merges. It is not a summarizer, topic grouper, protocol renderer or complete
+Canonical Meeting Knowledge stage.
+
+The first accepted V0 benchmark used `qwen3.5:9B` in one Ollama call over 33
+fact items. Runtime on the current machine was 390.119 seconds. One correct
+merge was accepted, involving `fact_0025` and `fact_0031`; 31 facts remained
+singletons, validation passed, no source fact was lost or duplicated, and
+non-fact categories remained unchanged. This is a local benchmark, not a
+general hardware claim.
+
+Broader semantic consolidation remains planned. It should group content by
+topic, mark contradictions and uncertainty, separate durable information from
+transient discussion, and produce Canonical Meeting Knowledge. It does not
+directly write a protocol.
 
 Canonical Meeting Knowledge is the planned semantic intermediate model and
 future single source of truth. It should be structured, preferably JSON, and
@@ -164,8 +189,8 @@ language is requested.
 - Topic segmentation exists as prototype tooling, not a stable pipeline stage.
 - Extraction is still a combined current flow, even though separate extractors
   are the intended architecture.
-- Deterministic Canonicalizer is documented but not implemented.
-- Semantic Consolidator is documented but not implemented.
+- Semantic Consolidator V0 is implemented only for facts-only duplicate
+  detection.
 - Canonical Meeting Knowledge is documented but not implemented.
 - Final output views are documented but not implemented.
 - Most prompt files are placeholders except the common and decision prompts.
@@ -176,5 +201,7 @@ language is requested.
 Stabilize repeatable local extraction evaluation before broadening the pipeline:
 expand Gold Standard coverage by category, keep one-chunk extraction as the
 baseline, and use small prompt experiments with immediate non-regression checks.
-After extraction behavior is stable enough, implement the Deterministic
-Canonicalizer first, then the Semantic Consolidator with evidence retention.
+The next recommended evaluation step is to use the consolidated V0 result as
+input for the unchanged Working Protocol renderer and compare that output
+against the Working Protocol Synthesizer V0 baseline and the human reference
+protocol.