test : added unit tests for finding_intelligence confidence and severity helpers

[tmdeveloper007]

Closes #1245.

Summary of What Has Been Done:
Added comprehensive unit tests for 7 pure helper functions in backend/secuscan/finding_intelligence.py: _normalize_severity, _severity_rank, _source_quality, _fingerprint_score, _build_confidence_reason, _compute_confidence, and _sort_sources. Tests cover normal inputs, edge cases (None, empty), and boundary conditions.

Changes Made:

• Created testing/backend/unit/test_finding_intelligence_confidence.py with 34 test cases organized in pytest classes
• _normalize_severity: 3 tests (known severities, case insensitivity, unknown defaults)
• _severity_rank: 3 tests (rank order, unknown defaults, comparability)
• _source_quality: 5 tests (known sources, max of multiple, unknown defaults, empty, whitespace)
• _fingerprint_score: 6 tests (all match_strength variants)
• _build_confidence_reason: 5 tests (reason composition, singular/plural, occurrences, fingerprint)
• _compute_confidence: 8 tests (valid range, source/evidence/occurrence/validation/severity factors)
• _sort_sources: 4 tests (dedup, sort, whitespace, empty)

Impact it Made:

• Prevents regressions in confidence computation logic
• Documents expected behavior for future contributors
• Increases overall test coverage of the finding_intelligence module

Note: This task is being handled by tmdeveloper007 -- please assign to that account when picking it up.