fix(heatmap): detection rules layer uses absolute rule count, not relative max

Before: score = (rules/max_rules)*50 + (evaluated/rules)*50
  -> everything red because relative to the 1 technique with most rules

After: score = min(rules/4 * 100, 100)  — absolute thresholds
  0 rules  = gray  (not covered)
  1 rule   = red   (25 — minimal)
  2 rules  = orange (50 — some)
  3 rules  = yellow (75 — good)
  4+ rules = green  (100 — well covered)

Also update HeatmapLegend labels to show actual rule counts instead of
meaningless percentage ranges.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

frontend/src/components/heatmap/HeatmapLegend.tsx

@@ -29,11 +29,11 @@ const LEGENDS: Record<
   "detection-rules": {
     label: "Detection Rules Coverage",
     colors: [
-      { color: "#d3d3d3", label: "No Rules (0)" },
-      { color: "#ff6666", label: "Few Rules (<25)" },
-      { color: "#ff9933", label: "Some Rules (25-50)" },
-      { color: "#ffff66", label: "Good Coverage (50-75)" },
-      { color: "#66ff66", label: "Full Coverage (75-100)" },
+      { color: "#d3d3d3", label: "0 rules" },
+      { color: "#ff6666", label: "1 rule" },
+      { color: "#ff9933", label: "2 rules" },
+      { color: "#ffff66", label: "3 rules" },
+      { color: "#66ff66", label: "4+ rules" },
     ],
   },
   campaign: {