Don't bother running realpath in DeviceUtils._GetChangedAndStaleFiles()

Don't bother running realpath in DeviceUtils._GetChangedAndStaleFiles()

Shaves off a single adb command from managed_install.py (~150ms)

BUG=520082
Committed: https://crrev.com/298f31eca43d36d67f6cbc8b2d2a495c5a625f1d
Cr-Commit-Position: refs/heads/master@{#345129}

[agrieve, 2015-08-20 01:09:12]

agrieve@chromium.org changed reviewers:
+ jbudorick@chromium.org

[agrieve, 2015-08-20 01:09:12]

Can't figure out why it was ever being run in the first place. Original CL
https://codereview.chromium.org/647013002

[jbudorick, 2015-08-23 02:19:57]

On 2015/08/20 at 01:09:12, agrieve wrote:
> Can't figure out why it was ever being run in the first place. Original CL
https://codereview.chromium.org/647013002

Its use predates that CL. IIRC, it was how we did things in AndroidCommands. I
don't remember why, though. If I had to guess, I'd say it's probably for the
sdcard, which tends to have some strange path things going on.

The logic here is definitely simpler, so if the bots are ok with it, then lgtm

[agrieve, 2015-08-24 17:50:35]

The CQ bit was checked by agrieve@chromium.org

[agrieve, 2015-08-24 17:50:36]

The patchset sent to the CQ was uploaded after l-g-t-m from
jbudorick@chromium.org
Link to the patchset: https://codereview.chromium.org/1300263002/#ps20001
(title: "rebase")

[commit-bot: I haz the power, 2015-08-24 17:50:54]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1300263002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1300263002/20001

[commit-bot: I haz the power, 2015-08-24 18:43:03]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-08-24 18:43:58]

Patchset 2 (id:??) landed as
https://crrev.com/298f31eca43d36d67f6cbc8b2d2a495c5a625f1d
Cr-Commit-Position: refs/heads/master@{#345129}

[agrieve, 2015-08-26 13:55:29]

A revert of this CL (patchset #2 id:20001) has been created in
https://codereview.chromium.org/1306883004/ by agrieve@chromium.org.

The reason for reverting is: md5sum prints realpath, so this is actually still
necessary unless we change md5sum to print paths given..