Add webhook failure email (and admin?) notice(s)

[thomasplevy]

If a webhook is disabled due to the maximum number of consecutive failures having been reached:

lifterlms-rest/includes/abstracts/class-llms-rest-webhook-data.php

Lines 289 to 321 in 9215df2

	protected function set_delivery_failure() {
	$failures = absint( $this->get( 'failure_count' ) );
	$this->set( 'failure_count', ++$failures );
	/**
	* Filter the number of times a webhook is allowed to fail before it is automatically disabled.
	*
	* @since 1.0.0-beta.1
	*
	* @param int $num Number of allowed failures. Default: 5.
	*/
	$max_allowed = apply_filters( 'llms_rest_webhook_max_delivery_failures', 5 );
	if ( $failures > $max_allowed ) {
	$this->set( 'status', 'disabled' );
	/**
	* Fires immediately after a webhook has been disabled due to exceeding its maximum allowed failures.
	*
	* @since 1.0.0-beta.1
	*
	* @param int $webhook_id ID of the webhook.
	*/
	do_action( 'llms_rest_webhook_disabled_by_delivery_failures', $this->get( 'id' ) );
	}
	return $this;
	}

There is currently no way for the admin to be aware of this other than proactively manually reviewing the webhooks list on the admin dashboard.

We should add an email notification that alerts users whenever a webhook fails. I think that creating this as a configurable email notification is probably best so that users can easily customize the behavior without code. Although the easiest solution might be to just make a simple email that can be customized with filters.

The notification should be sent to the webhook owner (and possibly to the site's main admin email address). Configuring this as an email notification makes it easy to add additional subscribers beyond the webhook owner too.

We could also pop an admin notice too?

This hook can be used to trigger:

lifterlms-rest/includes/abstracts/class-llms-rest-webhook-data.php

Line 315 in 9215df2

do_action( 'llms_rest_webhook_disabled_by_delivery_failures', $this->get( 'id' ) );

But it would probably be better to actually create a new hook that triggers on every failure (not just on disabled).

Ideally, I think, a notification should be triggered after 2 failures: first failure might be a momentary interruption, second and beyond failures probably mean there's a problem...