Test default compare in global status page

[thopre]

Summary

Make run for profile report older so that the global status page uses
another run.

[thopre]

Forgot to test without the change and realized today the test still passes so it's not testing the change.

[thopre]

Fix testcase to fail without D69056 and pass with

[thopre]

Ping?

[thopre]

In D87235#2294916, @thopre wrote:

Ping?

Ping @cmatthews ?

[cmatthews]

LGTM. Thanks!

[thopre]

In D87235#2309296, @cmatthews wrote:

LGTM. Thanks!

Thanks for the review! Landed as 22f39a4879da0908ac0abb0697aa5e73a9771934

tests/server/ui/Inputs/last-run-report.json

+{
+  "generated_by": "LNT Server v0.4.2.dev0",
+  "machine": {
+    "id": 4,
+    "name": "e105293.local__clang_DEV__x86_64"
+  },
+  "run": {
+    "start_time": "2020-09-11T12:01:10",
+    "end_time": "2020-09-11T12:02:19",
+    "order_by": "llvm_project_revision",
+    "llvm_project_revision": "154338",
+    "tag": "nts"
+  },
+  "tests": [
+    {
+      "name": "unique-name-wreodfpj",
+      "compile_status": "0",
+      "compile_time": "1.3",
+      "execution_status": "0",
+      "execution_time": "1.4",
+      "hash_status": "0",
+      "hash": "fedcba",
+      "score": 18,
+      "code_size": 10126
+    },
+    {
+      "name": "unique-name-lowrf",
+      "compile_status": "0",
+      "compile_time": "1.3",
+      "execution_status": "0",
+      "execution_time": "1.4",
+      "hash_status": "0",
+      "hash": "fedcba",
+      "score": 18,
+      "code_size": 10126
+    }
+  ]
+}

tests/server/ui/V4Pages.py

 from __future__ import print_function
 # Perform basic sanity checking of the V4 UI pages.
 #
 # create temporary instance
 # Cleanup temporary directory in case one remained from a previous run - also
 # see PR9904.
 # RUN: rm -rf %t.instance
 # RUN: python %{shared_inputs}/create_temp_instance.py \
 # RUN:   %s %{shared_inputs}/SmallInstance %t.instance \
 # RUN:   %S/Inputs/V4Pages_extra_records.sql
 
 # Import a profile
 # RUN: lnt import %t.instance %{shared_inputs}/profile-report.json
 #
+# Import last run. The run in this report must be the most recent one and the
+# tests inside unique to test the comparator used when sorting the tests per
+# regression status in the global status page.
+# RUN: lnt import %t.instance %S/Inputs/last-run-report.json
+#
 # RUN: lnt import %t.instance %S/Inputs/sample-failed-report1.json
 # RUN: lnt import %t.instance %S/Inputs/sample-failed-report2.json
 #
 # RUN: python %s %t.instance %{tidylib}
 
 from future import standard_library
 standard_library.install_aliases()
 import logging