Separate Task/Queue for SearchIndex #59

[jobelenus]

As discussed here: #59

Developed support for a specific Search Index to use a specific task path in order to leverage celery's queue routing to better manage resources (e.g. the number of workers) to complete certain real time indexes fast enough.

I also added the django app conf, since thats the new way to do things with the app loader. I also added Task class support of IGNORE_RESULTS via a setting.

[jobelenus]

Hrmm I saw the error:

flake8 celery_haystack --ignore=E501
celery_haystack/apps.py:1:1: F401 'CeleryHaystack' imported but unused
celery_haystack/apps.py:1:33: E261 at least two spaces before inline comment

Obviously the comment flake can be fixed -- but what about the unused import? AFAIK we do actually have to import that for the settings to get populated. What change would you prefer me to make so it passes travis?

[travijuu]

@jobelenus this is from flake documentation. this may help for the warnings?

lines that contain a # noqa comment at the end will not issue warnings