fix: add validation before queuing auto-retry cron jobs

[GaryJones]

Summary

• Adds validation checks before scheduling auto-retry cron jobs after pull failures
• Prevents indefinite requeuing of invalid jobs that can never complete successfully
• Refactors to use early returns for cleaner code flow

Problem

When a pull failure occurred, the auto-retry handler would schedule a cron job without validating the site data. This caused problems when:

1. The syn_site post no longer exists (get_post() returns null, causing fatal errors)
2. The post exists but is the wrong post type
3. The site has no valid syndication URL configured

These conditions would cause the cron job to keep requeuing indefinitely whilst never being able to complete successfully.

Solution

Added three validation checks to handle_pull_failure_event():

1. Post existence check: Verify the post exists and is a WP_Post instance
2. Post type check: Verify the post type is syn_site
3. URL validation: Verify the feed URL is valid using filter_var()

Also:

• Refactored to use early returns for cleaner code flow
• Changed date() to gmdate() to address PHPCS timezone warnings

Test plan

• All 91 unit tests pass
• New tests cover all validation scenarios (21 tests added)
• PHP syntax validated

Fixes #101

🤖 Generated with Claude Code