docs(cookbook): add missing values strategy and imputation cookbook guide #513

Context

As identified in the Growth & Developer Onboarding Audit, developers searching for practical solutions like "how to handle missing values pandas" or "safe tabular imputation" benefit from problem-oriented cookbook recipes in docs/cookbook/.

Task

Create a new problem-oriented cookbook guide in docs/cookbook/imputation-strategies-guide.md illustrating how FreshData approaches missing value imputation safely:

1. Explain the difference between simple univariate imputation (mean, median, mode) vs iterative model-based imputation (missforest).
2. Show how to configure per-column imputation rules vs global strategy.
3. Demonstrate FreshData's safety guarantees (e.g. refusal to impute declared ID or target columns).
4. Show how to inspect the generated CleanReport audit log to see exactly which rows and columns were modified.
5. Register the new recipe under the Cookbook section in mkdocs.yml.

Acceptance Criteria

• Create docs/cookbook/imputation-strategies-guide.md with complete markdown explanations and runnable Python examples.
• Document parameter combinations (strategy="conservative" vs "balanced" vs "aggressive", impute=...).
• Add the page to the nav list in mkdocs.yml under Cookbook.
• Verify that documentation builds cleanly without broken links using mkdocs build --strict.

Pointers

• Existing cookbook examples: docs/cookbook/
• Documentation config: mkdocs.yml
• Imputation engine reference: src/freshdata/engine/missing.py

New to the codebase? Comment /assign to claim this issue! You can also check our First Contribution Guide.