Bug 1317880 - Move adjust and thirdparty jars into GECKOVIEW_JARS to fix annotation processing r=sebastian GECKOVIEW_JARS contains two Adjust classes (within constants.jar). These depend on the rest of adjust, which was previously within FENNEC_JARS. Without the remaining adjust jars being on the classpath during annotation processing for GECKOVIEW_JARS, we are unable to correctly process those Adjust classes (i.e. we get a NoClassDefFoundError). The minimal fix is to process adjust as part of GECKOVIEW_JARS. Because adjust depends on HttpClientLib, we also need to move the relevant jars into GECKOVIEW_JARS too (sync-thirparty contains HttpClientLib). This will probably require further untangling, this is a minimal patch to allow beta to actually build. MozReview-Commit-ID: DLtazTrg3hV

Bug 1317347 - Fix exception in network monitor for requests coming from iframes added to the hidden window. r=jryans MozReview-Commit-ID: eRVaHTbyx1

Bug 1317570 - Add a mochitest to test 'ended' fires only once when seeking to the end. r=cpearce MozReview-Commit-ID: KRm1shND6lL

Bug 1317576. Part 7 - remove the assertion that doesn't make sense anymore. r=cpearce It is possible to call AudioEndTime() in CompletedState::Step() with MedisSink not started at all. This happens when seeking to the end on a video only file. MozReview-Commit-ID: 78P72DAPpWI

Bug 1317576. Part 6 - ensure 'playbackEnded' is notified when seeking to the end on a paused media element. r=cpearce MozReview-Commit-ID: 4fzru2oHaPV

Bug 1317576. Part 5 - run Step() in CompletedState::Enter() without scheduling an additional cycle. r=cpearce MDSM might have reached the end of playback. We want to notify 'playbackEnded' ASAP. MozReview-Commit-ID: FYI6xTWuNSD

Bug 1317576. Part 4 - remove unused MediaDecoder::SeekResolveValue. r=cpearce MozReview-Commit-ID: 9H6e0gFvWeT

Bug 1317576. Part 3 - remove unused parameter/member. r=cpearce MozReview-Commit-ID: Ff6a8MKVY8q

Bug 1317576. Part 2 - always transition to DECODING when seek is done. r=cpearce DecodingState::Enter() will check whether decoding is completed and transition to COMPLETED. MozReview-Commit-ID: 5abPWWulGWo

Bug 1317576. Part 1 - add some helper functions. r=cpearce MozReview-Commit-ID: FNWmQohsDRq

Bug 1317587 - don't start scheduled sync after shutdown. r=rnewman MozReview-Commit-ID: GSyWAUa1qyr

Backed out changeset 21b69215ab80 (bug 1316870) for forthcoming merge conflicts

Bug 1317631 - Post: add more cursor position sanity checks r=sebastian MozReview-Commit-ID: 7CBnLZDkExI

Bug 1317631 - Only handle clicks for highlights r=sebastian This avoids us trying to obtain an invalid cursor position, since the cursor only maps to highlights items (and not the headers). MozReview-Commit-ID: 1NtJuvDRa5r

Bug 1315391 - Clean up actor destruction after changing to `destroy`. r=ochameau The `destroy` method in some actors would throw errors or was incomplete, leading to various issues closing the toolbox after the previous patch. This cleans up all such cases noticed through manual testing of the toolbox. MozReview-Commit-ID: 6EZYFwjSri

Bug 1315391 - Rename all `disconnect` methods to `destroy` in actors. r=ochameau Ever since protocol.js was added as a way to create DevTools actors, we've had lots of confusion about the correct way to implement actor destruction. If your actor's _parent_ was the legacy kind, you had to use `disconnect`. If it was protocol.js, you had to use `destroy`. There is no reason for this madness, which makes reasoning about destruction quite hard. Here we rename `disconnect` to `destroy` so there is only one name for every destruction path. MozReview-Commit-ID: C1Yw9NfUUR2

Bug 1316870 - Enable no-shadow eslint rule for browser/. r=mossop MozReview-Commit-ID: 17vLN6W4BVj

Bug 1313788 - Remove DOM/wifi and related code. r=jst MozReview-Commit-ID: BRT6bSvPpT

Bug 1317822 - Move GMPCrashHelper into its own file. r=gerald MozReview-Commit-ID: 7CinZ2Y2Fmz

Bug 1317592 - don't try syncing other engines after one fails with a shutdown error. r=rnewman MozReview-Commit-ID: LZ72JbOtpm4