855 monitor intersection movements for missing high volume movements

[gabrielwol]

What this pull request accomplishes:

• adds a view to find common movements in the last 100 days which are not in intersection_movements table
  • add a weekly notification to check this view (see example in #airflow_pipelines-dev)
• Adds a denylist: intersection_movements_denylist with same structure as base ("allowlist")
  • Didn't want to add this as column to existing table because it would require a lot more join changes
• Adds BEFORE INSERT triggers on both intersection_movements and intersection_movements_denylist to restrict inserting overlapping rows between the two. See it in action below.
• Note there is a new after insert trigger (to insert padding values based on new inserts to intersection_movements) being developed in Miovision Aggregation Improvements (ATR view, intersection_movements AFTER INSERT trigger, unacceptable_gaps) #1024 which will make it easier to make additions to intersection_movements.

Test new triggers using:

INSERT INTO miovision_api.intersection_movements
SELECT * FROM miovision_api.intersection_movements_denylist LIMIT 1;

INSERT INTO miovision_api.intersection_movements_denylist
SELECT * FROM miovision_api.intersection_movements LIMIT 1;

Issue(s) this solves:

• Closes Need to monitor intersection_movements for missing high volume movements #855

What, in particular, needs to reviewed:

• The two functions are basically the same, maybe I should make them take a table parameter... ✅

What needs to be done by a sysadmin after this PR is merged

E.g.: these tables need to be migrated/created in the production schema.

[chmnata]

just some questions on the view, possibility of using 1 function for two tables 🤔

[chmnata]

I think we are missing a where im_dl.blahblah is null in the where clause for the anti join

[chmnata]

when I select from the view, there was one row where sum_volume > intersection_classification_total which might be sus.

[gabrielwol]

Not what I exepcted! It's because SELECT now()::date AT TIME ZONE 'Canada/Eastern' - interval ' 100 days' results in 2024-05-21 20:00:00 so the larger value had an extra 4 hrs of data.

[chmnata]

is it possible to make the table name a param?

[gabrielwol]

Done!

[chmnata]

:gabe-approved: