perf(tools): lower default result sizes to cut response tokens

The default limits let broad queries flood the model context. This
change sets smaller defaults. An explicit parameter still overrides each one.

- file_search: max_results 1000 to 50, context_lines 2 to 0.
- ast_query: max_results 100 to 50.
- ast_query compact output omits the node type unless verbose is set.

On a fixed seven-call suite, response size fell from 198529 to 61960 bytes.

internal/query/query.go

@@ -509,11 +509,15 @@ func formatResults(results []MatchResult, maxResults int, format string, offset
 	case "compact":
 		for i := 0; i < renderCount; i++ {
 			r := results[i]
+			firstLine := firstLineOf(r.Text, 80)
+			if !verbose {
+				fmt.Fprintf(&sb, "%s:%d %s\n", r.File, r.Location.Line, firstLine)
+				continue
+			}
 			nodeType := "unknown"
 			if r.Node != nil {
 				nodeType = r.Node.Type()
 			}
-			firstLine := firstLineOf(r.Text, 80)
 			fmt.Fprintf(&sb, "%s:%d (%s) %s\n", r.File, r.Location.Line, nodeType, firstLine)
 		}