Refactor ReportGenerator to modularize HTML and SARIF report generation

[RohitKattimani]

Description

Currently, the ReportGenerator class in backend/secuscan/reporting.py contains monolithic methods, specifically _generate_pdf_html_report, generate_html_report, and generate_sarif_report. These methods handle data parsing, conditional logic, and large string concatenations all in a single block.

This structure makes the file difficult to maintain and goes against the project's backend code style guideline to "prefer small functions over large, multi-purpose blocks".

Proposed Solution

I propose refactoring these methods by breaking them down into smaller, isolated helper methods (e.g., _build_pdf_finding_markup, _build_web_finding_markup, and _extract_sarif_rule_id).

• The underlying behavior, functionality, and generated outputs of the reports will not change.
• The refactor will strictly focus on improving code readability, testability, and modularity.
• I will ensure the backend test suite (./testing/test_python.sh) passes locally before submitting the PR.

Additional Context

I am participating in GSSoC 2026 and would love to work on this! Please assign this issue to me, and I will open a Pull Request with the refactored code shortly.

Best,
Rohit Kattimani

[RohitKattimani]

@utksh1 You may review and assign me this issue