docs: clarify security issue classification

[HTHou]

Summary

• classify untrusted-file findings according to exploitability and demonstrated security-boundary impact
• clarify that crashes, assertion failures, out-of-memory conditions, and sanitizer findings without such impact are normally robustness issues
• retain case-by-case security treatment for disproportionate resource amplification across an actual untrusted boundary
• narrow the examples of security-relevant memory findings to exploitable corruption or out-of-bounds access that exposes unrelated process data

This follow-up refines the security model added in #936 and more clearly reflects the distinction used by comparable Apache data-format projects.

Testing

• git diff --check
• documentation-only change; no code tests required

[ColinLeeo]

LGTM

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 61.67%. Comparing base (b6ed43c) to head (2a5af68).
⚠️ Report is 1 commits behind head on develop.

Additional details and impacted files

@@           Coverage Diff            @@
##           develop     #937   +/-   ##
========================================
  Coverage    61.67%   61.67%           
========================================
  Files          747      747           
  Lines        51883    51883           
  Branches      8437     8437           
========================================
  Hits         32000    32000           
  Misses       18255    18255           
  Partials      1628     1628           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.