feat: Complete markdown syntax refinement and variable protection

✅ New cleaner syntax implemented:
- [card:type] instead of [card type='type']
- [button:style](url)Text[/button] instead of [button url='...' style='...']
- Standard markdown images: ![alt](url)

✅ Variable protection from markdown parsing:
- Variables with underscores (e.g., {order_items_table}) now protected
- HTML comment placeholders prevent italic/bold parsing
- All variables render correctly in preview

✅ Button rendering fixes:
- Buttons work in Visual mode inside cards
- Buttons work in Preview mode
- Button clicks prevented in visual editor
- Proper styling for solid and outline buttons

✅ Backward compatibility:
- Old syntax still supported
- No breaking changes

✅ Bug fixes:
- Fixed order_item_table → order_items_table naming
- Fixed button regex to match across newlines
- Added button/image parsing to parseMarkdownBasics
- Prevented button clicks on .button and .button-outline classes

📚 Documentation:
- NEW_MARKDOWN_SYNTAX.md - Complete user guide
- MARKDOWN_SYNTAX_AND_VARIABLES.md - Technical analysis

admin-spa/src/lib/html-to-markdown.ts

@@ -0,0 +1,64 @@
+/**
+ * Convert HTML to Markdown
+ * Simple converter for rich text editor output
+ */
+
+export function htmlToMarkdown(html: string): string {
+  if (!html) return '';
+  
+  let markdown = html;
+  
+  // Headings
+  markdown = markdown.replace(/<h1>(.*?)<\/h1>/gi, '# $1\n\n');
+  markdown = markdown.replace(/<h2>(.*?)<\/h2>/gi, '## $1\n\n');
+  markdown = markdown.replace(/<h3>(.*?)<\/h3>/gi, '### $1\n\n');
+  markdown = markdown.replace(/<h4>(.*?)<\/h4>/gi, '#### $1\n\n');
+  
+  // Bold
+  markdown = markdown.replace(/<strong>(.*?)<\/strong>/gi, '**$1**');
+  markdown = markdown.replace(/<b>(.*?)<\/b>/gi, '**$1**');
+  
+  // Italic
+  markdown = markdown.replace(/<em>(.*?)<\/em>/gi, '*$1*');
+  markdown = markdown.replace(/<i>(.*?)<\/i>/gi, '*$1*');
+  
+  // Links
+  markdown = markdown.replace(/<a\s+href="([^"]+)"[^>]*>(.*?)<\/a>/gi, '[$2]($1)');
+  
+  // Lists
+  markdown = markdown.replace(/<ul[^>]*>(.*?)<\/ul>/gis, (match, content) => {
+    const items = content.match(/<li[^>]*>(.*?)<\/li>/gis) || [];
+    return items.map((item: string) => {
+      const text = item.replace(/<li[^>]*>(.*?)<\/li>/is, '$1').trim();
+      return `- ${text}`;
+    }).join('\n') + '\n\n';
+  });
+  
+  markdown = markdown.replace(/<ol[^>]*>(.*?)<\/ol>/gis, (match, content) => {
+    const items = content.match(/<li[^>]*>(.*?)<\/li>/gis) || [];
+    return items.map((item: string, index: number) => {
+      const text = item.replace(/<li[^>]*>(.*?)<\/li>/is, '$1').trim();
+      return `${index + 1}. ${text}`;
+    }).join('\n') + '\n\n';
+  });
+  
+  // Paragraphs - convert to double newlines
+  markdown = markdown.replace(/<p[^>]*>(.*?)<\/p>/gis, '$1\n\n');
+  
+  // Line breaks
+  markdown = markdown.replace(/<br\s*\/?>/gi, '\n');
+  
+  // Horizontal rules
+  markdown = markdown.replace(/<hr\s*\/?>/gi, '\n---\n\n');
+  
+  // Remove remaining HTML tags
+  markdown = markdown.replace(/<[^>]+>/g, '');
+  
+  // Clean up excessive newlines
+  markdown = markdown.replace(/\n{3,}/g, '\n\n');
+  
+  // Trim
+  markdown = markdown.trim();
+  
+  return markdown;
+}