feat(llm): add LLM integration module with OpenRouter and Claude Code adapters

Implements markitect/llm/ package with concrete LLMAdapter implementations:
- OpenRouterAdapter: HTTP via urllib with retry/backoff on 429/5xx
- ClaudeCodeAdapter: subprocess-based Claude CLI with stdin piping
- Factory pattern: create_adapter("openrouter") or create_adapter("claude-code")
- API key resolution chain: constructor > env var > project-root key file
- 42 unit tests, 2 integration tests (gated on API key / CLI availability)

Also adds the infospace-with-history example with Wealth of Nations VSM
analysis pipeline, templates, schemas, source chapters, and processed
output for chapters 1-2. process_chapters.py now supports --provider
and --model flags for automatic LLM-driven processing.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

examples/infospace-with-history/output/metrics/metrics-report.md

@@ -0,0 +1,152 @@
+# Infospace Metrics Report
+
+**Report Date:** 2026-02-10
+**Chapters Processed:** 1 of 35
+
+---
+
+## Completeness Metrics
+
+### VSM Concept Coverage
+
+**Coverage: 5 of 12 concepts (41.7%)**
+
+| Concept | Covered | Mapping Count |
+|---------|---------|---------------|
+| S1 (Operations) | Yes | 8 |
+| S2 (Coordination) | Yes | 3 |
+| S3 (Control) | No | 0 |
+| S3* (Audit) | No | 0 |
+| S4 (Intelligence) | Yes | 2 |
+| S5 (Policy) | No | 0 |
+| Recursion | Yes | 1 |
+| Variety | No | 0 |
+| Requisite Variety | No | 0 |
+| Attenuation/Amplification | No | 0 |
+| Algedonic Signals | No | 0 |
+| Viability | Yes | 1 |
+| Autonomy | No | 0 |
+
+**Uncovered concepts:** S3, S3*, S5, Variety, Requisite Variety,
+Attenuation/Amplification, Algedonic Signals, Autonomy
+
+**Assessment:** Coverage is concentrated on S1 (Operations), which is expected
+for the opening chapter focused on production. The remaining concepts require
+chapters addressing regulation (S3), policy (S5), and information management
+(S3*, variety engineering).
+
+### Chapter Coverage
+
+**Coverage: 1 of 35 chapters (2.9%)**
+
+| Book | Chapters Available | Chapters Processed |
+|------|-------------------|-------------------|
+| Introduction | 1 | 0 |
+| Book I | 11 | 1 |
+| Book II | 6 | 0 |
+| Book III | 4 | 0 |
+| Book IV | 10 | 0 |
+| Book V | 3 | 0 |
+
+### Entity Count
+
+**Total distinct entities: 13**
+
+| Economic Domain | Count |
+|----------------|-------|
+| Production | 10 |
+| Distribution | 1 |
+| Exchange | 1 |
+| General Theory | 1 |
+| Consumption | 0 |
+| Accumulation | 0 |
+| Regulation | 0 |
+
+**Assessment:** Entity extraction is heavily skewed toward Production, reflecting
+the chapter's content. Domains like Accumulation and Regulation will require
+Book II and Book V chapters respectively.
+
+### Mapping Count
+
+**Total mappings: 14**
+
+Mapping strength distribution:
+- Strong: 11 (78.6%)
+- Moderate: 3 (21.4%)
+- Weak: 0 (0.0%)
+
+---
+
+## Consistency Metrics
+
+### Terminology Consistency
+
+**Score: 1.0**
+
+With only one chapter processed, no cross-chapter terminology conflicts exist.
+All entity names follow Smith's own terminology and the naming conventions
+specified in the extraction rules. No synonyms detected.
+
+**Flagged issues:** None
+
+### Cross-reference Integrity
+
+**Score: 1.0**
+
+All entity references in mapping documents point to entities defined in the
+entities output. No broken references detected.
+
+**Flagged issues:** None
+
+### Schema Compliance
+
+**Compliance: 100%**
+
+| Document Type | Total | Compliant | Non-compliant |
+|--------------|-------|-----------|---------------|
+| Entity documents | 13 | 13 | 0 |
+| Mapping documents | 14 | 14 | 0 |
+| Analysis documents | 1 | 1 | 0 |
+
+All documents contain their required sections as defined by their respective
+schemas.
+
+---
+
+## Recommendations
+
+### Priority Actions
+
+1. **Process remaining Book I chapters** to build S2 and S3 coverage through
+   price theory (Chapters 5-7) and wage/profit regulation (Chapters 8-10).
+
+2. **Process Book IV chapters** for S5 (Policy) coverage, as these address
+   political economy systems (mercantilism, free trade).
+
+3. **Process Book V chapters** for S3 (Control) coverage through sovereign
+   revenue and public administration.
+
+### Coverage Gaps to Watch
+
+- **Variety and requisite variety** are core VSM concepts that may not map
+  directly to Smith's vocabulary. Consider whether Smith's discussions of
+  market size, competition, and information asymmetry can be mapped to
+  variety management.
+
+- **Algedonic signals** may appear in discussions of famine, market crises,
+  or economic distress in later chapters.
+
+- **S3* (Audit)** is the most difficult system to map in classical economics.
+  Watch for discussions of market inspections, trade regulation enforcement,
+  or quality verification.
+
+### Quality Notes
+
+- Current consistency scores (1.0 across all metrics) will be the baseline.
+  As more chapters are processed, terminology conflicts and cross-reference
+  issues are likely to emerge and should be actively managed.
+
+- The high proportion of Strong mappings (78.6%) is appropriate for Chapter 1
+  given its focus on operations, which map cleanly to S1. Later chapters
+  covering more abstract economic concepts may yield more Moderate and Weak
+  mappings.