Update redis to fix celery worker/beat connectivity issue when kombu is upgraded

[pkfec]

While working on #3566, discovered that the scheduled tasks and cron jobs stopped running in dev and stage environments. The celery worker and beat apps on dev and stage environments failed to start during circle builds and also manual restart of those apps did work either. The status of these two apps remained in CRASHED status and non of these two app instances were UP and RUNNING.

After taking a close look at the celery worker logs, it appeared that the kombu AMQP(Advanced Messaging Queue Protocol) is expecting a higher version of redis than what we have currently in openFEC/requirements.txt.

Here is the log stack trace from celery worker app:
2019-03-06T14:01:56.64-0500 [APP/PROC/WEB/0] ERR kombu.exceptions.VersionMismatch: Redis transport requires redis-py versions 3.2.0 or later. You have 2.10.6

Problem:
Sprint8.2 release cut happened on Mar 5, 2019. During the build, celery4.1.1 grabbed the newer version of kombu >=4.2.0,<5.0. The kombu package was updated on Mar 3, 2019 and we are not aware of this change. The newer version of kombu4.4.0 is compatible with redis3.2.0 only.
Kombu :https://github.com/celery/celery/blob/master/requirements/default.txt
Redis :https://github.com/celery/celery/blob/master/requirements/extras/redis.txt

Our development stack has redis 2.10.6 which is not compatible with the latest kombu4.4.0. This caused the celery worker/beat apps CRASH in dev and stage environments.

Fix/Resolution:
Celery worker/beat app were UP and RUNNING after upgrading redis version to 3.2.0 and deploy the changes to dev and stage spaces.

Redis upgrade work is implemented in this PR #3605

[pkfec]

The fix is deployed to dev and stage spaces already. Fix will be deployed toprod during the release-20190312.

Here is the PR to track the changes : #3605

Closing this issue.