Add benchmark that imitates news reading.

tools/perf/page_sets/data/browsing_mobile.json

Index: tools/perf/page_sets/data/browsing_mobile.json
diff --git a/tools/perf/page_sets/data/browsing_mobile.json b/tools/perf/page_sets/data/browsing_mobile.json
new file mode 100644
index 0000000000000000000000000000000000000000..37b649cd0125f2b62ab5ebd75f2beb3f25d3ee03
--- /dev/null
+++ b/tools/perf/page_sets/data/browsing_mobile.json
@@ -0,0 +1,29 @@
+{
+    "archives": {
+        "browsing_mobile_000.wpr": [
+            "browse:news:cnn"
+        ],
+        "browsing_mobile_001.wpr": [
+            "browse:social:facebook"
+        ],
+        "browsing_mobile_002.wpr": [
+            "browse:news:hackernews"
+        ],
+        "browsing_mobile_003.wpr": [
+            "browse:news:nytimes"
+        ],
+        "browsing_mobile_004.wpr": [
+            "browse:news:qq"
+        ],
+        "browsing_mobile_005.wpr": [
+            "browse:news:reddit"
+        ],
+        "browsing_mobile_006.wpr": [
+            "browse:social:twitter"
+        ],
+        "browsing_mobile_007.wpr": [
+            "browse:news:washingtonpost"
+        ]
+    },
+    "description": "Describes the Web Page Replay archives for a story set. Don't edit by hand! Use record_wpr for updating."
+}