rustdoc: Omit repeated paths in the search index.

Since the items roughly follow the lexical order, there are
many consecutive items with the same path value which can be
easily compressed.

For the library and compiler docs, this commit decreases
the index size by 26% and 6% before and after gzip, respectively.

src/librustdoc/html/static/main.js

@@ -539,7 +539,7 @@
 
                 // an array of [(Number) item type,
                 //              (String) name,
-                //              (String) full path,
+                //              (String) full path or empty string for previous path,
                 //              (String) description,
                 //              (optional Number) the parent path index to `paths`]
                 var items = rawSearchIndex[crate].items;
@@ -561,10 +561,11 @@
                 // all other search operations have access to this cached data for
                 // faster analysis operations
                 var len = items.length;
+                var lastPath = "";
                 for (var i = 0; i < len; i += 1) {
                     var rawRow = items[i];
                     var row = {crate: crate, ty: rawRow[0], name: rawRow[1],
-                               path: rawRow[2], desc: rawRow[3],
+                               path: rawRow[2] || lastPath, desc: rawRow[3],
                                parent: paths[rawRow[4]]};
                     searchIndex.push(row);
                     if (typeof row.name === "string") {
@@ -573,6 +574,7 @@
                     } else {
                         searchWords.push("");
                     }
+                    lastPath = row.path;
                 }
             }
             return searchWords;