Delete unused FilterURLTermiate_* actions.

Delete unused FilterURLTermiate_* actions.

These actions currently account for about 5% of Android
UMA data (most taken by FilterURLTermiate_About) and also:
a) do not have any owners, b) have typo in the name and
c) are not even actual user actions (i.e. not directly user
initiated).

If this data is interesting to someone, the metrics can
be re-added with a histogram, which doesn't carry timing
info and thus would be much more efficient.

BUG=387734
Committed: https://crrev.com/f65686cdf2a36ae94d4a39dbf5ec8d50ff7c8e4f
Cr-Commit-Position: refs/heads/master@{#328393}

[Alexei Svitkine (slow), 2015-05-05 17:08:59]

asvitkine@chromium.org changed reviewers:
+ avi@chromium.org, mpearson@chromium.org

[Alexei Svitkine (slow), 2015-05-05 17:09:00]

avi: content/
mpearson: tools/metrics/

[Mark P, 2015-05-05 17:13:19]

metrics lgtm

[Avi (use Gerrit), 2015-05-05 17:18:37]

avi@chromium.org changed reviewers:
+ nasko@chromium.org

[Avi (use Gerrit), 2015-05-05 17:18:37]

These were added by Nasko in https://chromiumcodereview.appspot.com/12208056 two
years ago. Nasko?

[nasko, 2015-05-05 17:27:23]

On 2015/05/05 17:18:37, Avi wrote:
> These were added by Nasko in https://chromiumcodereview.appspot.com/12208056
two
> years ago. Nasko?

The data is interesting, but having a more efficient way to collect it will be
useful. Any pointers to documentation on what the efficiency of different
approaches will be highly appreciated.

Since I don't plan on turning FilterURL failures into process kills at this
time, it is ok to remove those for now. LGTM

[Alexei Svitkine (slow), 2015-05-05 17:32:51]

On 2015/05/05 17:27:23, nasko wrote:
> On 2015/05/05 17:18:37, Avi wrote:
> > These were added by Nasko in https://chromiumcodereview.appspot.com/12208056
> two
> > years ago. Nasko?
> 
> The data is interesting, but having a more efficient way to collect it will be
> useful. Any pointers to documentation on what the efficiency of different
> approaches will be highly appreciated.
> 
> Since I don't plan on turning FilterURL failures into process kills at this
> time, it is ok to remove those for now. LGTM

We have some internal docs at go/uma-metrics - i.e. you can read about
histograms vs. user actions - but generally you can think of a histogram as
client aggregated whereas actions are not - so histograms are more efficient
(but do not carry timing information - something actions have but is usually
unused since our dashboards don't show it).

[Alexei Svitkine (slow), 2015-05-05 17:33:52]

The CQ bit was checked by asvitkine@chromium.org

[commit-bot: I haz the power, 2015-05-05 17:36:57]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1125893003/1

[commit-bot: I haz the power, 2015-05-05 20:29:28]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-05-05 20:30:25]

Patchset 1 (id:??) landed as
https://crrev.com/f65686cdf2a36ae94d4a39dbf5ec8d50ff7c8e4f
Cr-Commit-Position: refs/heads/master@{#328393}