feat(extractor): implement multi-engine article content extractor

- Added scripts/extract_article_contents.py for batch scraping with stealth Foxcape and triple extraction (Trafilatura, Newspaper4k, Readability)
- Created unit, integration, and E2E test suite in tests/test_extract_article_contents.py (90/90 passing)
- Updated specs/003-article-content-extractor and README.md with usage documentation and CLI contracts
- Passed ruff linting/formatting and mypy type checking cleanly

tests/test_adversarial.py

@@ -7,9 +7,8 @@ and CLI execution behavior via subprocess (exit codes, stream purity, JSON parsi
 import json
 import subprocess
 import sys
-from pathlib import Path
-import pytest
-from src.models import ECPSnapshot, RelatedEntity, DecisionCategory
+
+from src.models import DecisionCategory, ECPSnapshot, RelatedEntity
 
 
 def test_adversarial_sao_paulo_city_vs_fc():
@@ -20,9 +19,13 @@ def test_adversarial_sao_paulo_city_vs_fc():
         aliases=["São Paulo", "SPFC", "Tricolor Paulista"],
         domain="Futebol e Esportes",
         anchors=["Morumbi", "futebol", "campeonato", "Copa Libertadores", "elenco", "estádio"],
-        negative_anchors=["prefeitura de são paulo", "governo do estado de são paulo", "trânsito na capital paulista"],
+        negative_anchors=[
+            "prefeitura de são paulo",
+            "governo do estado de são paulo",
+            "trânsito na capital paulista",
+        ],
         graph_version="1.0.0",
-        related_entities=[]
+        related_entities=[],
     )
     content = (
         "# Obras Viárias na Capital\n\n"
@@ -30,6 +33,7 @@ def test_adversarial_sao_paulo_city_vs_fc():
         "para desafogar o fluxo de veículos na região central durante os horários de pico."
     )
     from src.classifier import InherenceClassifier
+
     classifier = InherenceClassifier()
     result = classifier.classify(ecp, content)
     assert result.decision in (DecisionCategory.NOT_RELATED, DecisionCategory.TANGENTIAL)
@@ -47,13 +51,14 @@ def test_adversarial_apple_fruit_recipe():
         anchors=["iPhone", "MacBook", "iOS", "silicon", "hardware"],
         negative_anchors=["apple pie", "orchard harvest", "doce de maçã"],
         graph_version="1.0.0",
-        related_entities=[]
+        related_entities=[],
     )
     content = (
         "# Receita Caseira\n\n"
         "Comprei maçãs frescas no mercado para preparar um doce de maçã com canela e açúcar mascavo."
     )
     from src.classifier import InherenceClassifier
+
     classifier = InherenceClassifier()
     result = classifier.classify(ecp, content)
     assert result.decision in (DecisionCategory.NOT_RELATED, DecisionCategory.TANGENTIAL)
@@ -76,9 +81,9 @@ def test_adversarial_related_entity_without_scope_context():
                 relation_type="SUPPLIER_OF",
                 weight=0.95,
                 scope="battery_technology",
-                confidence=0.99
+                confidence=0.99,
             )
-        ]
+        ],
     )
     # Content mentions Northvolt in an unrelated/passing architectural context without domain anchors
     content = (
@@ -87,6 +92,7 @@ def test_adversarial_related_entity_without_scope_context():
         "mit moderner Holzfassade und Blick auf den See."
     )
     from src.classifier import InherenceClassifier
+
     classifier = InherenceClassifier()
     result = classifier.classify(ecp, content)
     # Must be TANGENTIAL or NOT_RELATED, NEVER CONTEXTUAL_INHERENT
@@ -98,18 +104,27 @@ def test_adversarial_related_entity_without_scope_context():
 def test_adversarial_subprocess_cli_success_stdout(tmp_path):
     """Run CLI via subprocess without --output and verify stdout is pure parseable JSON."""
     ecp_file = tmp_path / "ecp.json"
-    ecp_file.write_text(json.dumps({
-        "target_entity_id": "ent_petrobras",
-        "target_name": "Petrobras",
-        "aliases": ["Petrobras"],
-        "domain": "Oil & Gas",
-        "anchors": ["petróleo", "pré-sal"]
-    }), encoding="utf-8")
+    ecp_file.write_text(
+        json.dumps(
+            {
+                "target_entity_id": "ent_petrobras",
+                "target_name": "Petrobras",
+                "aliases": ["Petrobras"],
+                "domain": "Oil & Gas",
+                "anchors": ["petróleo", "pré-sal"],
+            }
+        ),
+        encoding="utf-8",
+    )
 
     content_file = tmp_path / "content.md"
-    content_file.write_text("# Notícia\n\nA Petrobras bateu recorde de extração de petróleo no pré-sal este mês.", encoding="utf-8")
+    content_file.write_text(
+        "# Notícia\n\nA Petrobras bateu recorde de extração de petróleo no pré-sal este mês.",
+        encoding="utf-8",
+    )
 
     import os
+
     env = dict(os.environ, PYTHONIOENCODING="utf-8", PYTHONUTF8="1")
 
     res = subprocess.run(
@@ -133,16 +148,22 @@ def test_adversarial_subprocess_cli_success_stdout(tmp_path):
 def test_adversarial_subprocess_cli_empty_content(tmp_path):
     """Run CLI via subprocess with empty content and verify error code and exit code."""
     import os
+
     env = dict(os.environ, PYTHONIOENCODING="utf-8", PYTHONUTF8="1")
 
     ecp_file = tmp_path / "ecp.json"
-    ecp_file.write_text(json.dumps({
-        "target_entity_id": "ent_1",
-        "target_name": "Test",
-        "aliases": ["Test"],
-        "domain": "Tech",
-        "anchors": ["tech"]
-    }), encoding="utf-8")
+    ecp_file.write_text(
+        json.dumps(
+            {
+                "target_entity_id": "ent_1",
+                "target_name": "Test",
+                "aliases": ["Test"],
+                "domain": "Tech",
+                "anchors": ["tech"],
+            }
+        ),
+        encoding="utf-8",
+    )
 
     content_file = tmp_path / "empty.md"
     content_file.write_text("   \n\n  ", encoding="utf-8")
@@ -164,15 +185,21 @@ def test_adversarial_subprocess_cli_empty_content(tmp_path):
 def test_adversarial_subprocess_cli_missing_required_field(tmp_path):
     """Run CLI via subprocess with missing target_name and verify error payload."""
     import os
+
     env = dict(os.environ, PYTHONIOENCODING="utf-8", PYTHONUTF8="1")
 
     ecp_file = tmp_path / "ecp_bad.json"
-    ecp_file.write_text(json.dumps({
-        "target_entity_id": "ent_1",
-        "aliases": ["Test"],
-        "domain": "Tech",
-        "anchors": ["tech"]
-    }), encoding="utf-8")
+    ecp_file.write_text(
+        json.dumps(
+            {
+                "target_entity_id": "ent_1",
+                "aliases": ["Test"],
+                "domain": "Tech",
+                "anchors": ["tech"],
+            }
+        ),
+        encoding="utf-8",
+    )
 
     content_file = tmp_path / "content.md"
     content_file.write_text("Conteúdo de teste válido.", encoding="utf-8")
@@ -193,6 +220,7 @@ def test_adversarial_subprocess_cli_missing_required_field(tmp_path):
 def test_adversarial_subprocess_cli_corrupted_json(tmp_path):
     """Run CLI via subprocess with corrupted JSON and verify error payload."""
     import os
+
     env = dict(os.environ, PYTHONIOENCODING="utf-8", PYTHONUTF8="1")
 
     ecp_file = tmp_path / "ecp_corrupted.json"