* monitor "schemas" queue to run refresh_schema
`refresh_schema` tasks won't run because "schemas" queue isn't consumed
with default settings.
and it cause leaking redis storage, a "schemas" list is growing with time.
this PR fix it, adds "schemas" queue to monitor by celery.
* use scheduled_worker for "schemas" queue
instead of "adhoc_worker"
* open port 3000 for remote debugging
* add ptvsd
* use port 5678 to avoid changes in VSCode's default config
* attach to ptvsd
* no need to wait for attach
* actually, --debugger seems to be working
* create a new docker entry point for remote debugging
* alternative method to switch to debugging
* Fix tag counts for dashboards to be distinct.
This also makes use of the Dashboard.all base query.
Fix#3108.
* Use Query.all_queries as the base query for Query.all_tags.
* Add test case for Dashboard.all_tags.
* Update docker-compose.yml
jut updating docker-compose dev to version 3, any special reason to keep redis on 3? and pg on 9.5? I could also add a volume to pg, any reason not to do so?
* rollback to redis 3 and pg 9.5 due to consistency in project defaults
* Configure volume directly in worker service.
`restart: always` will cause containers to restart when the machine is rebooted or the docker process restarts. Setting this to `unless-stopped` will mean the container will restart only if the docker process/computer was rebooted without stopping the container first. Otherwise, running `docker-compose stop` and rebooting will cause the docker containers all to start again (and goodbye laptop battery!)
We might also consider `restart: on-failures` which would mean that rebooting would _not_ cause containers to start again unless requested via `docker-compose up`.
