If we fail to obtain the update center metadata, don't keep retrying
every 15 seconds, but use exponential time out to avoid overloading
our server in case of a problem.

done with post-production (as opposed to just done with build steps). Jobs
that had long running tasks that occured in their post-production state, i.e.,
build wrappers, post-build steps, could get into the situation where a queued
build would start running on an executor while the previous build was still
running post-build actions. If the two executors shared the same workspace,
the new build could delete information with a clean that the current build
still needed for processing.

- Changed AbstractProject.getCauseOfBlockage() to call isLogUpdated instead
  of isBuilding.

   core/src/main/java/hudson/model/AbstractProject.java

- Added isLogUpdated.

   core/src/main/java/hudson/model/Job.java

- Added test case.

   test/src/test/java/hudson/model/JobQueueTest.java

- Change log

   changelog.html

This hopefully avoids showing menus when the user didn't intend to see
them

Bundle slf4j binding to the war.
See the comment in war/pom.xml for detailed discussion.

This is fundamentally a "damned if I do, damned if I don't" situation,
but given that JENKINS-12334 is a fatal error, and the downside of
bundling the binding jar is "multiple binding" warning, it seems like
the lesser evil is to bundle it and risk some warnings.

If the sections unfold/fold etc and layout changes, the save button
needs to update its location