Improve output reports with more details

[sakost]

Summary

The current report outputs (text, JSON, HTML) provide basic mutation testing results but lack several useful details that would help users understand and act on the results.

Current State

• Text report: Shows aggregate stats (files scanned, mutants generated/tested/killed/survived/timeouts/errors) and lists survived mutants with file:line, operator name, and original → mutated text
• JSON report: Serializes the full MutationReport struct
• HTML report: Self-contained with summary table, per-file annotated source with color-coded lines (killed=green, survived=red, no-coverage=gray)

Suggested Improvements

Per-file summary

• Add a per-file breakdown showing mutation score per file
• Helps identify which modules have the weakest test coverage

Per-operator summary

• Show statistics grouped by mutation operator (e.g., "ArithmeticOp: 45 killed / 50 total = 90%")
• Helps identify which types of mutations are consistently surviving

Per-test coverage analysis

• Show which tests killed which mutants (data is available from pytest-cov context)
• Helps understand test effectiveness

Diff-style output

• Show mutations as unified diffs rather than just "original → mutated" text
• Easier to understand the context of each mutation

Comparative/baseline reports

• Support comparing results across runs (e.g., "score improved from 85% to 92%")
• Useful for CI integration

HTML report enhancements

• Show full source line content (currently only shows the mutated fragment)
• Add sortable/filterable tables
• Add a navigation sidebar for multi-file reports
• Add mutation score trend chart (when session DB is available)

Getting Started

The report implementations are in src/report/:

• text.rs — text report
• json.rs — JSON report
• html.rs — HTML report
• types.rs — MutationReport and MutantResult data structures

The MutationReport struct already contains all individual MutantResult entries with file paths, operator names, and status — most improvements just need to aggregate this existing data differently.

Pick any single improvement from the list above and submit a PR!