feat(context): add content line counts to AGENTS.md heading extraction

Replace extract_headings with extract_heading_summaries that returns Vec<(String, usize)>
with line counts for each top-level heading. Update resolve_agents_md to format summaries
as 'Heading (N lines), Heading2 (M lines)' format.

Verifies v2-phase5.AC1.1

src/context/agents_md.rs

···
             let agents_md_path = dir.join("AGENTS.md");
             if agents_md_path.exists() && !seen_paths.contains(&agents_md_path) {
                 seen_paths.insert(agents_md_path.clone());
-                let headings = extract_headings(&agents_md_path)?;
-                let heading_summary = headings.join(", ");
+                let heading_summaries = extract_heading_summaries(&agents_md_path)?;
+                let heading_summary = heading_summaries
+                    .iter()
+                    .map(|(heading, count)| format!("{} ({} lines)", heading, count))
+                    .collect::<Vec<_>>()
+                    .join(", ");
                 let path_str = agents_md_path.to_string_lossy().to_string();
                 summaries.push((path_str, heading_summary));
             }
···
     Ok(summaries)
 }
 
-/// Extract top-level headings (lines starting with "# ") from a markdown file
-fn extract_headings(path: &Path) -> Result<Vec<String>> {
+/// Extract top-level headings with content line counts from a markdown file
+///
+/// Returns a vector of (heading_text, line_count) tuples, where line_count is the number
+/// of non-empty content lines under that heading (until the next heading or EOF).
+fn extract_heading_summaries(path: &Path) -> Result<Vec<(String, usize)>> {
     let content = std::fs::read_to_string(path)?;
-    let mut headings = Vec::new();
+    let lines: Vec<&str> = content.lines().collect();
+    let mut summaries = Vec::new();
+
+    let mut i = 0;
+    while i < lines.len() {
+        let line = lines[i];
 
-    for line in content.lines() {
+        // Check if this is a top-level heading (# followed by space)
         if let Some(heading) = line.strip_prefix("# ") {
-            headings.push(heading.trim().to_string());
+            let heading_text = heading.trim().to_string();
+
+            // Count non-empty lines until the next top-level heading or EOF
+            let mut line_count = 0;
+            let mut j = i + 1;
+            while j < lines.len() {
+                let next_line = lines[j];
+                // Stop at the next top-level heading
+                if next_line.starts_with("# ") {
+                    break;
+                }
+                // Count non-empty lines
+                if !next_line.trim().is_empty() {
+                    line_count += 1;
+                }
+                j += 1;
+            }
+
+            summaries.push((heading_text, line_count));
         }
+
+        i += 1;
     }
 
-    Ok(headings)
+    Ok(summaries)
 }
 
 #[cfg(test)]
···
     use tempfile::TempDir;
 
     #[test]
-    fn test_extract_headings() -> Result<()> {
+    fn test_extract_heading_summaries() -> Result<()> {
+        let tmpdir = TempDir::new()?;
+        let agents_md_path = tmpdir.path().join("AGENTS.md");
+        fs::write(
+            &agents_md_path,
+            "# Introduction\nSome content here\nMore content\n# Getting Started\n\n## Subsection\ndetails\n# Advanced",
+        )?;
+
+        let summaries = extract_heading_summaries(&agents_md_path)?;
+        assert_eq!(summaries.len(), 3);
+        assert_eq!(summaries[0].0, "Introduction");
+        assert_eq!(summaries[0].1, 2); // "Some content here" and "More content"
+        assert_eq!(summaries[1].0, "Getting Started");
+        assert_eq!(summaries[1].1, 2); // "## Subsection" and "details" (both non-empty content lines)
+        assert_eq!(summaries[2].0, "Advanced");
+        assert_eq!(summaries[2].1, 0); // no content after
+        Ok(())
+    }
+
+    #[test]
+    fn test_extract_heading_summaries_empty_sections() -> Result<()> {
         let tmpdir = TempDir::new()?;
         let agents_md_path = tmpdir.path().join("AGENTS.md");
         fs::write(
             &agents_md_path,
-            "# Introduction\n# Getting Started\n## Subsection\n# Advanced",
+            "# First\n\n\n# Second\nContent\n# Third\n",
         )?;
 
-        let headings = extract_headings(&agents_md_path)?;
-        assert_eq!(
-            headings,
-            vec!["Introduction", "Getting Started", "Advanced"]
-        );
+        let summaries = extract_heading_summaries(&agents_md_path)?;
+        assert_eq!(summaries.len(), 3);
+        assert_eq!(summaries[0].1, 0); // First has no content
+        assert_eq!(summaries[1].1, 1); // Second has one line
+        assert_eq!(summaries[2].1, 0); // Third has no content
         Ok(())
     }
 
···
         let project_root = tmpdir.path();
 
         // Create AGENTS.md at root
-        fs::write(project_root.join("AGENTS.md"), "# Root\n# Guidelines")?;
+        fs::write(project_root.join("AGENTS.md"), "# Root\nroot content\n# Guidelines")?;
 
         // Create a file to scope
         fs::write(project_root.join("main.rs"), "fn main() {}")?;
 
         let summaries = resolve_agents_md(project_root, &[PathBuf::from("main.rs")])?;
         assert_eq!(summaries.len(), 1);
-        assert_eq!(summaries[0].1, "Root, Guidelines");
+        assert_eq!(summaries[0].1, "Root (1 lines), Guidelines (0 lines)");
         Ok(())
     }