Security fixes

• Upgrade Tor Browser to 5.0.3. (Closes: #10223)

• Upgrade bind9-based packages to 1:9.8.4.dfsg.P1-6+nmu2+deb7u7.

• Upgrade liblcms1 to 1.19.dfsg2-1.2+deb7u1.

• Upgrade libldap-2.4-2 to 2.4.31-2+deb7u1.

• Upgrade libslp1 to 1.2.1-9+deb7u1.

• Upgrade ssl-cert to 1.0.32+deb7u1.

• Bugfixes

• Fix a corner case for the MAC spoofing panic mode. If panic mode
  failed to disable the specific device that couldn't be spoofed
  (by unloading the module) we disable networking. Previously we
  only stopped NetworkManager. The problem is that NM isn't even
  started at this time, but will specifically be started when
  we're done with MAC spoofing. Therefore, let's completely
  disable NetworkManager so it cannot possibly be
  started. (Closes: #10160)

• Avoid use of uninitialized value in restricted-network-detector.
  If NetworkManager decides that a wireless connection has timed
  out before "supplicant connection state" has occued, our idea of
  the state is undef, so it cannot be used in a string
  comparison. Hence, let's initialize the state to the empty
  string instead of undef. Also fix the state
  recording. Apparently NetworkManager can say a few different
  things when it logs the device state transitions. (Closes:

  7689)

• Minor improvements

• Remove workaround for localizing search engine plugins. The
  workaround has recently become unnecessary, possibly due to the
  changes made for the seach bar after the Tor Browser was rebased
  on Firefox 38esr. (Closes: #9146)

• Refer to the I2P Browser in the I2P notifications. Instead of
  some obscure links that won't work in the Tor Browser, where
  users likely will try them, and which I believe will open them
  by default. (Closes: #10182)

• Upgrade I2P to 0.9.22. Also set the I2P apparmor profile to
  enforce mode. (Closes: #9830)

• Test suite

• Test that udev-watchdog is monitoring the correct device when
  booted from USB. (Closes: #9890)

• Remove unused 'gksu' step. This causes a false-positive to be
  found for #5330. (Closes: #9877)

• Make --capture capture individual videos for failed scenarios
  only, and --capture-all to capture videos for all scenarios.
  (Closes: #10148)

• Use the more efficient x264 encoding when capturing videos using
  the --capture* options. (Closes: #10001)

• Make --old-iso default to --iso if omitted. Using the same ISO
  for the USB upgrade tests most often still does what we want,
  e.g. test that the current version of Zero Trace Pen being tested has a
  working Zero Trace Pen installer. Hence this seems like a reasonable
  default. (Closes: #10147)

• Avoid nested FindFailed exceptions in waitAny()/findAny(), and
  throw a new dedicated FindAnyFailed exception if these fail
  instead. Rjb::throw doesn't block Ruby's execution until the
  Java exception has been received by Ruby, so strange things can
  happen and we must avoid it. (Closes: #9633)

• Fix the Download Management page in our browsers. Without the
  browser.download.panel.shown pref set, the progress being made
  will not update until after the browser has been restarted.
  (Closes: #8159)

• Add a 'pretty_debug' (with an alias: 'debug') Cucumber formatter
  that deals with debugging instead of printing it to STDERR via
  the --debug option (which now has been removed). This gives us
  the full flexibility of Cucumber's formatter system, e.g. one
  easy-to-read formatter can print to the terminal, while we get
  the full debug log printed to a file. (Closes: #9491)

• Import logging module in otr-bot.py. Our otr-bot.py does not use
  logging but the jabberbot library makes logging calls, causing a
  one-off message “No handlers could be found for logger
  "jabberbot"” to be printed to the console. This commit
  effectively prevents logging/outputting anything to the terminal
  which is at a level lower than CRITICAL. (Closes: 9375)

• Force new Tor circuit and reload web site on browser
  timeouts. (Closes: #10116)

• Focus Pidgin's buddy list before trying to access the tools
  menu. (Closes: #10217)

• Optimize IRC test using waitAny. If connecting to IRC fails,
  such as when OFTC is blocking Tor, waiting 60 seconds to connect
  while a a Reconnect button is visible is sub-optimal. It would
  be better to try forcing a new Tor circuit and clicking the
  reconnect button. (Closes: #9653)

• Wait for (and focus if necessary) Pidgin's Certificate windows.
  (Closes: #10222)