Drop "prior work"/"proof-of-concept" hedging; fix stale 313+/5-discipline stat

"Prior Work & Research Contributions" -> "Role & Contributions". The
overview and closing lines described the Framework as prior work / a
proof-of-concept -- reframed as what it actually is: the live,
domain-agnostic methodology behind CopernicusAI's process analysis,
already running GLMP and ATAP.

Also fixed the "Scale Demonstration" stat, which was badly stale --
"313+ processes across 5 disciplines (..., Mathematics: 20, ...)" versus
the actual live count (692 across 6 families, Mathematics/ATAP: 226 --
pre-dated the ATAP expansion by a wide margin). Verified fresh against
knowledge-engine-status.json's process_databases field before writing
the new numbers in.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

index.html

@@ -59,12 +59,15 @@
     <!-- Prior Work & Research Contributions -->
     <section class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8 py-12">
         <div class="bg-gradient-to-r from-orange-50 to-yellow-50 rounded-xl shadow-lg p-8 mb-8">
-            <h2 class="text-3xl font-bold text-gray-900 mb-6">📚 Prior Work & Research Contributions</h2>
-            
+            <h2 class="text-3xl font-bold text-gray-900 mb-6">📚 Role & Contributions</h2>
+
             <div class="bg-white rounded-lg p-6 mb-6">
                 <h3 class="text-xl font-semibold text-gray-900 mb-4">Overview</h3>
                 <p class="text-gray-700 mb-4">
-                    The Programming Framework represents <strong>prior work</strong> that demonstrates a novel methodology for analyzing complex processes by combining Large Language Models (LLMs) with visual flowchart representation. This research establishes a universal, domain-agnostic approach to process analysis that transforms textual descriptions into structured, interactive visualizations.
+                    The Programming Framework is the domain-agnostic methodology behind CopernicusAI's process
+                    analysis — combining Large Language Models (LLMs) with visual flowchart representation to
+                    transform textual process descriptions into structured, interactive visualizations across any
+                    discipline.
                 </p>
             </div>
 
@@ -76,7 +79,7 @@
                         <li>• <strong>LLM-Powered Extraction:</strong> Automated extraction using Google Gemini 2.0 Flash</li>
                         <li>• <strong>Structured Visualization:</strong> Mermaid.js-based flowchart generation encoded as JSON</li>
                         <li>• <strong>Iterative Refinement:</strong> Systematic approach enabling continuous improvement</li>
-                        <li>• <strong>Scale Demonstration:</strong> Applied to 313+ processes across 5 disciplines (Biology: 52, Chemistry: 91, Physics: 21, Computer Science: 21, Mathematics: 20, GLMP: 109)</li>
+                        <li>• <strong>Scale Demonstration:</strong> Applied to 692 processes across 6 families (GLMP: 217, Mathematics/ATAP: 226, Biology: 55, Chemistry: 123, Physics: 21, Computer Science: 50)</li>
                         <li>• <strong>Validation:</strong> Successfully processes complex biological, chemical, and computational workflows with high accuracy</li>
                     </ul>
                 </div>
@@ -109,7 +112,9 @@
                     </ul>
                 </div>
                 <p class="text-gray-600 text-sm italic">
-                    This work establishes a proof-of-concept for AI-assisted process analysis, demonstrating how LLMs can systematically extract and visualize complex logic from textual sources across diverse domains.
+                    This is the foundational methodology for AI-assisted process analysis — LLMs systematically
+                    extract and visualize complex logic from textual sources across diverse domains, from GLMP's
+                    regulatory circuits to ATAP's proof structures.
                 </p>
             </div>
         </div>