Page experience: FID issue possibly caused by @ampproject/pull/34028

Details
We had gotten a report in slack about a possible FID regression caused by the `img` changes in https://github.com/ampproject/amphtml/pull/34028
---

Hello,
I work on a few sister sites that share the same code base and AMP template. Since May 28th they have been experiencing Core Web Vitals issues on AMP pages, specifically First Input Delay. However, we haven’t made any code changes to AMP recently and these are blog pages there are no interactive features that should trigger FID to be measured.
The only user interactive function I can think of is that images on AMP have a lightbox feature but this is bundled into AMP by default. I checked the AMPHTML changelogs and there is a change to AMP images and lightbox at the end of April. CrUX data has a 28-day collection period so seeing the issues being reported at the end of May lines up with this change.
Can any AMP devs comment on the possibility this change being related to FID issues? and would be able to investigate/report it to the appropriate channel?

---

Components in use:
"amp-bind", "amp-ad", "amp-analytics", "amp-social-share", "amp-iframe", "amp-youtube", "amp-twitter", "amp-instagram", "amp-facebook", "amp-brightcove", "amp-sidebar", "amp-sticky-ad", "amp-video", "amp-fx-collection", "amp-auto-lightbox", "amp-lightbox-gallery"

/cc @ampproject/wg-performance