fix: task descriptions not showing for specs with compact markdown

Tasks like spec 053 were missing descriptions in the kanban board and
task modal because the regex for extracting the overview from spec.md
required two newlines after "## Overview" (a blank line).

Specs generated with compact markdown (no blank line after headers)
were not matched, resulting in empty descriptions.

Changes:
- Fix regex to accept one or more newlines: /## Overview\s*\n+/
- Add fallback to read from requirements.json task_description field
- Extract meaningful content from GitHub issue descriptions

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

auto-claude-ui/src/main/project-store.ts

@@ -243,8 +243,8 @@ export class ProjectStore {
         if (existsSync(specFilePath)) {
           try {
             const content = readFileSync(specFilePath, 'utf-8');
-            // Extract first paragraph after "## Overview"
-            const overviewMatch = content.match(/## Overview\s*\n\n([^\n#]+)/);
+            // Extract first paragraph after "## Overview" - handle both with and without blank line
+            const overviewMatch = content.match(/## Overview\s*\n+([^\n#]+)/);
             if (overviewMatch) {
               description = overviewMatch[1].trim();
             }
@@ -258,6 +258,42 @@ export class ProjectStore {
           description = plan.description;
         }
 
+        // Fallback: read description from requirements.json if still not found
+        if (!description) {
+          const requirementsPath = path.join(specPath, AUTO_BUILD_PATHS.REQUIREMENTS);
+          if (existsSync(requirementsPath)) {
+            try {
+              const reqContent = readFileSync(requirementsPath, 'utf-8');
+              const requirements = JSON.parse(reqContent);
+              if (requirements.task_description) {
+                // Extract a clean summary from task_description (first line or first ~200 chars)
+                const taskDesc = requirements.task_description;
+                const firstLine = taskDesc.split('\n')[0].trim();
+                // If the first line is a title like "Investigate GitHub Issue #36", use the next meaningful line
+                if (firstLine.toLowerCase().startsWith('investigate') && taskDesc.includes('\n\n')) {
+                  const sections = taskDesc.split('\n\n');
+                  // Find the first paragraph that's not a title
+                  for (const section of sections) {
+                    const trimmed = section.trim();
+                    // Skip headers and short lines
+                    if (trimmed.startsWith('#') || trimmed.length < 20) continue;
+                    // Skip the "Please analyze" instruction at the end
+                    if (trimmed.startsWith('Please analyze')) continue;
+                    description = trimmed.substring(0, 200).split('\n')[0];
+                    break;
+                  }
+                }
+                // If still no description, use a shortened version of task_description
+                if (!description) {
+                  description = firstLine.substring(0, 150);
+                }
+              }
+            } catch {
+              // Ignore parse errors
+            }
+          }
+        }
+
         // Try to read task metadata
         const metadataPath = path.join(specPath, 'task_metadata.json');
         let metadata: TaskMetadata | undefined;