Delay shutting down secondary transport when device app is backgrounded

[NicoleYarroch]

Fixes #1560

This PR is ready for review.

Risk

This PR makes no API changes.

Testing Plan

• I have verified that I have not introduced new warnings in this PR (or explain why below)
• I have run the unit tests with this PR
• I have tested this PR against Core and verified behavior (if applicable, if not applicable, explain why below).

Unit Tests

Unit tests were added to the SecondaryTransportManagerSpec.

Core Tests

Tests were performed with a video streaming app using a TCP secondary transport. Tests were considered successful:

1. If the background task did not expire before the app was foregrounded then video resumed and audio continued without interruption.
2. If the the background task expired while the app was backgrounded, then when the app was brought back to the foreground video and audio resumed.

Tests Performed

1. Put app on device in the background and brought app on device back to foreground before background task expired.
2. Put app on device in the background and brought app on device back to foreground after the background task expired.
3. Put app on device in the background and then locked the device. Unlocked the device before the background task expired and foregrounded the app.
4. Put app on device in the background and then locked the device. Waited for the background task to expire. Unlocked the device and foregrounded the app.
5. Put app on device in the background and then locked the device. Waited for the background task to expire. Attempted to launch the SDL app while the device was still locked, the attempting to connect alert shows up. Unlocked the device and foregrounded the app.
6. Locked the device while the app on the device was still in foreground. Unlocked the device.
7. Put app on device in background, opened another SDL video streaming app, then switched back to the app and brought it back to the foreground.

Core version / branch / commit hash / module tested against: SYNC 4 (20016_DEVTEST)
HMI name / version / branch / commit hash / module tested against: SYNC 4 (20016_DEVTEST)

Summary

When the app on the device is backgrounded the secondary transport is now closed when the background task expires instead of immediately. This improves user experience by:

1. Avoiding a reconnect delay if the user backgrounds the app and then foregrounds the app within the time the background task is kept alive (~1 minute).
2. Showing the videoStreamBackgroundString when the app on the device is backgrounded. Previously the warning message was either flashed for a moment or not shown all.

In order to ensure that the video, audio and tcp transports are all shutdown before the background task is ended, I moved calling the videoServiceEndedCompletionHandler and the audioServiceEndedCompletionHandler from the stop service ACK and NAK handler methods to the stopped state of the SDLStreamingVideoLifecycleManager and SDLStreamingAudioLifecycleManager to ensure that both managers' properties get reset before the background task ends when the app is backgrounded. This also reduced duplicate code since both the ACK/NAK handlers transition to the stopped state anyways.

Changelog

Enhancements

• When the app on the device is backgrounded the secondary transport is now closed when the background task expires instead of immediately.

CLA

• I have signed the CLA

[joeljfischer]

While testing, I saw a few more issues:

1. I consistently saw a background task ended that was unknown:

2020-04-14 14:44:03.388147-0400 SDL Example[1175:377941] Can't end BackgroundTask: no background task exists with identifier 3 (0x3), or it may have already been ended. Break in UIApplicationEndBackgroundTaskError() to debug.

It doesn't seem to be causing an issue, but it does make me a bit suspicious.

[NicoleYarroch]

While testing, I saw a few more issues:

1. I consistently saw a background task ended that was unknown:

2020-04-14 14:44:03.388147-0400 SDL Example[1175:377941] Can't end BackgroundTask: no background task exists with identifier 3 (0x3), or it may have already been ended. Break in UIApplicationEndBackgroundTaskError() to debug.

It doesn't seem to be causing an issue, but it does make me a bit suspicious.

I saw the same notifications and it seems to be a known iOS 13 issue. I created an empty single-view Swift app and I get the exact same log Can't end BackgroundTask: no background task exists with identifier 2 (0x2), or it may have already been ended. Break in UIApplicationEndBackgroundTaskError() to debug. every time I foreground the app after backgrounding it.

[joeljfischer]

Please see the remaining request for comment addition.