Refactor: Break reporting.py into rendering, artifact, naming, and diff services

[techmore]

Title: Break reporting.py into rendering, artifact, naming, and diff services

Type: refactor
Severity: medium
Area: backend / reporting / report lifecycle

Description:
reporting.py has become a broad backend module that mixes XML parsing, asset extraction, diff generation, HTML/PDF rendering, persisted artifact metadata, customer naming normalization, saved-scan lookup, and task-oriented helper behavior. That concentration increases regression risk in a core subsystem that already affects history, reports, downloads, and exports.

Evidence:

• /Users/techmore/projects/NmapUI/nmapui/reporting.py is roughly 1400+ lines
• it contains multiple distinct responsibilities including:
  • report artifact path resolution
  • customer identity normalization
  • XML asset parsing and diff summarization
  • scan metadata persistence
  • HTML/PDF conversion helpers
  • saved-scan lookup and PDF regeneration paths

Proposed Fix:
Split reporting into smaller backend services with stable interfaces.

Implementation Notes:

• likely extraction candidates:
  • report_artifacts.py
  • report_rendering.py
  • report_diffs.py
  • report_identity.py
• keep a thin orchestration layer for workflows so task code does not import internals directly
• add targeted module tests before moving logic to preserve current runtime behavior

Related Issues:

• Initiative: Report rendering parity and PDF stylesheet strategy #106
• Bug: Saved reports are named after auto-detect network instead of the resolved public-IP customer identity #132
• Refactor: Add schema versioning and migrations for the SQLite runtime store #142

[techmore]

Tracking under #166 (Reporting module modularization).