feat: index the text of attached PDFs

- jz_bm_material_file_texts caches the pdftotext output per file
- the River job jz.bm.extract_material_text extracts and resyncs
- the management multipart attach queues it inside its transaction
- file_texts in the document and a file snippet with its name

plugin_test.go

@@ -79,8 +79,8 @@ func TestPluginActivates(t *testing.T) {
 	if n := len(p.Models()); n != 14 {
 		t.Errorf("Models = %d, want 14", n)
 	}
-	if n := len(p.Migrations()); n != 19 {
-		t.Errorf("Migrations = %d, want 19", n)
+	if n := len(p.Migrations()); n != 20 {
+		t.Errorf("Migrations = %d, want 20", n)
 	}
 	if !slices.ContainsFunc(p.Commands(), func(c bonfire.Command) bool { return c.Name == "bm:seed" }) {
 		t.Error("bm:seed is not registered")