BUG: Page parentage update fix.

[mfendeksilverstripe]

SiteTree re-organization fix

The condition did not properly match sometimes due to the type comparison which ended up causing unwanted purge static cache jobs.

Workaround for silverstripe/silverstripe-versioned#360

Split off from #132

[dhensby]

@mfendeksilverstripe can you make sure you open PRs from forks in the future, please.

As for this change, we might as well use loose compactors rather than explicit casting in this case? Or is there some benefit to explicitly casting as ints? As a personal peev, I dislike the assumption that DB IDs should be ints because it really ties our hands in the future

[michalkleiner]

I'd second Dan's opinion here, allowing string IDs in future to support e.g. UUIDs instead of sequential numeric IDs will need to be a thing, so using loose comparison here would be OK from my perspective, or, if we have to have a strict comparison, going the other way and comparing it as strings.

[mfendeksilverstripe]

Switched to loose comparison as per your recommendation @michalkleiner .

cc @dhensby