feat(anomaly): Implement RelativeRateOfChangeStrategy for anomaly detection

[ericpsimon]

Summary

Implements a new anomaly detection strategy pattern for the Term data validation library, providing flexible rate-of-change based anomaly detection with asymmetric thresholds and robust edge case handling.

Closes TER-167

What does this PR do?

• Introduces AnomalyDetectionStrategy trait as an alternative to the existing detector pattern
• Implements RelativeRateOfChangeStrategy with configurable asymmetric thresholds
• Adds comprehensive input validation and floating-point edge case handling
• Provides 10 comprehensive tests covering all edge cases
• Includes working example demonstrating usage patterns

Type of change

• New feature (non-breaking change which adds functionality)
• Bug fix (non-breaking change which fixes an issue)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update

Key Features

• Asymmetric Thresholds: Different limits for increases vs decreases (e.g., allow 20% growth but only 5% decline)
• Robust Validation: Rejects NaN, infinity, and negative rate thresholds at construction
• Epsilon-based Comparisons: Handles floating-point precision issues correctly
• Zero Baseline Handling: Special handling for near-zero values to prevent division errors
• Configurable History Requirements: Set minimum history points needed for detection
• Rich Result Details: Provides confidence scores, expected ranges, and detailed explanations

Checklist

• Tests pass locally (10 tests added, all passing)
• Code formatted with cargo fmt
• No clippy warnings (cargo clippy -- -D warnings)
• Documentation updated with rustdoc comments
• Example added (examples/src/anomaly_detection_strategy.rs)
• Input validation with comprehensive error messages
• Edge cases handled (NaN, infinity, near-zero, negative values)

Testing

# Run strategy tests
cargo test -p term-guard --lib analyzers::anomaly::strategy::tests

# Run example
cargo run --example anomaly_detection_strategy

Performance Considerations

• O(1) algorithm examining only the last historical point
• Minimal allocations in hot path
• Pre-allocated HashMap capacity where possible
• Zero-copy patterns for historical data

🤖 Generated with Claude Code

Co-Authored-By: Claude noreply@anthropic.com