docs: comprehensive audit and DX overhaul of all reference modules

llms.md:
- Only Groq/OpenAI/LiteLLM/HuggingFaceLLM are exported — remove non-exported
  Anthropic/Ollama/Gemini/DeepSeek/Novita as direct imports
- Rename HuggingFace -> HuggingFaceLLM (correct class name)
- Remove non-existent create_provider() — replace with LiteLLM provider/model pattern
- Add LiteLLM 100+ providers section with provider/model string examples
- Add Exported Classes table (class -> provider -> API key)
- Update Provider Comparison table to show correct import per provider

ontology.md:
- Remove non-existent OntologyManager — replace with OntologyEngine facade
- Remove non-existent start_explorer() — replace with CLI: semantica-explorer
- SHACLValidator -> OntologyValidator (correct exported name)
- OWLExporter -> OWLGenerator (correct exported name)
- Add Exported Classes block with all 15+ exported symbols
- Add LLMOntologyGenerator section, NamespaceManager section
- Add OntologyEvaluator section with coverage/completeness metrics
- Add ingest_ontology() section
- Add versioning moved-to note (change_management module)

kg.md:
- TemporalKnowledgeGraph does not exist — replace with TemporalGraphQuery
- DistanceCalculator does not exist — replace with SimilarityCalculator
- Add Exported Classes block with all 20+ exported symbols
- Fix temporal example to use TemporalGraphQuery + TemporalVersionManager correctly
- Add SimilarityCalculator section with NodeEmbedder integration example

provenance.md:
- ActivityTracker not exported — remove; ProvenanceManager handles tracking
- Fix track_entity() signature: add source_location, source_quote params
- Fix GraphBuilderWithProvenance import: from semantica.kg, not semantica.provenance
- Add Exported Classes block with storage backends and checksum utilities
- Add SourceReference section with DOI/page/quote fields
- Add tamper-evident checksum section (compute_checksum/verify_checksum)
- Add Enable Provenance in Extractors section
- Fix duplicate heading (W3C PROV-O Export appeared twice)

reasoning.md:
- Add Exported Classes block with all engines + data types + explanation types
- Add Quick Start section
- Add Choosing an Engine comparison table
- Add InferenceResult/Explanation/ReasoningStep type annotations in examples
- Add Tip: use DatalogReasoner for recursive rules

semantic_extract.md:
- Add Exported Classes block with NamedEntityRecognizer, EventDetector, Entity,
  Relation, Event, CoreferenceChain, EntityClassifier, TemporalEventProcessor
- Add Quick Start section (one-liner extraction pipeline)
- Rename EventExtractor -> EventDetector (correct exported name)
- Clarify NERExtractor vs NamedEntityRecognizer distinction
- Add return type annotations to EventDetector example

core.md:
- Add Exported Classes block
- Add When to Use Core vs. Individual Modules decision table
- Add Tip: LifecycleManager only for long-running apps
- Fix MethodRegistry example to import build_knowledge_base correctly

parse.md:
- Add Exported Classes block with all format-specific parsers + data types
- Add DoclingParser optional import note

utils.md:
- Add Exported Classes block with logging/validation/progress/helpers/exceptions

deduplication.md:
- Add Exported Classes block with PropertyMergeRule, MergeStrategyManager,
  method_registry, and all convenience functions

export.md:
- Add Exported Classes block with all exporters, NamespaceManager,
  SemanticNetworkYAMLExporter, and all convenience functions

docs/reference/utils.md

@@ -6,6 +6,31 @@ icon: "wrench"
 
 `semantica.utils` provides shared infrastructure used throughout Semantica. Most users won't call it directly, but its APIs are available when you need fine-grained control over logging, validation, progress tracking, or error handling.
 
+## Exported Classes
+
+```python
+from semantica.utils import (
+    # Logging
+    setup_logging,    # configure root logger — level, format (json/text)
+    get_logger,       # get a named logger instance
+    log_performance,  # @decorator — logs function name, duration, exception
+    # Validation
+    validate_entity,  # validate entity dict structure, raises ValidationError
+    validate_config,  # validate config dict against schema, raises ValidationError
+    # Progress tracking
+    ProgressTracker,  # class-based tracker with ETA
+    track_progress,   # wraps any iterable with live progress bar
+    # Helpers
+    clean_text,       # normalize whitespace, strip control characters
+    hash_data,        # deterministic SHA-256 hash of any serializable object
+    safe_filename,    # sanitize a string for use as a filename
+    # Exceptions
+    SemanticaError,   # base exception for all Semantica errors
+    ValidationError,  # raised when input fails validation
+    ProcessingError,  # raised during extraction, graph build, or pipeline step
+)
+```
+
 ## What You Get
 
 <CardGroup cols={2}>