tools/perf: Run non-abridged system health on FYI

We're running system health benchmarks on FYI bots to support
development of experimental TBMv3 metrics. We realized the abridged
version didn't include any accessiblity story, so we could not test the
accessibility metric. In general, since we're interested in making sure
TBMv3 metrics work well across a wide range of scenarios, it's best to
turn on the non-abridged version so we can detect any strange errors
early on.

If the load on the bots proves to be too much we can rollback and
manually pick a better subset of stories.

Change-Id: I7c2538b36b41b414a33ade5749a2953603602914
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2388262Reviewed-by: John Chen <johnchen@chromium.org>
Commit-Queue: Deep Roy <dproy@google.com>
Cr-Commit-Position: refs/heads/master@{#803600}

tools/perf/core/shard_maps/linux-perf-fyi_map.json

@@ -5,7 +5,7 @@
                 "abridged": false
             },
             "system_health.common_desktop": {
-                "abridged": true
+                "abridged": false
             }
         }
     },

tools/perf/core/shard_maps/win-10_laptop_low_end-perf_hp-candidate_map.json

@@ -5,7 +5,7 @@
                 "abridged": false
             },
             "system_health.common_desktop": {
-                "abridged": true
+                "abridged": false
             }
         }
     },