Enhance loop analysis with sophisticated detection algorithms

Major improvements to clustrix/loop_analysis.py:

**Enhanced Parallelizability Assessment:**
- More sophisticated iteration count analysis (minimum 3 iterations)
- Dependency checking for external variables
- Pattern recognition for NumPy/Pandas operations
- Conservative approach for reliable parallelization decisions

**New Loop Analysis Methods:**
- estimate_parallelization_benefit(): Scores loops 0.0-1.0 for benefit
- suggest_parallelization_strategy(): Recommends executor type and chunking
- analyze_loop_patterns(): Comprehensive analysis with performance estimates

**Enhanced Dependency Analysis:**
- Array access pattern detection (read/write conflicts)
- Reduction operation identification (+=, *=, etc.)
- Control flow statement detection (break, continue, return)
- Loop-carried dependency analysis for safety

**New Capabilities:**
- Vectorization opportunity detection
- Performance estimation with confidence levels
- Strategic recommendations for parallel execution
- Detailed analysis of nested loops and patterns

**Maintained Compatibility:**
- All existing tests pass (36/36)
- Backward compatible API preserved
- Type hints improved for better development experience

Resolves TODO at line 58 and implements sophisticated analysis algorithms
requested in GitHub issue #46.

🤖 Generated with [Claude Code](https://claude.ai/code)

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

Author: jeremymanning