Better handle html parsing in diverse backoffice inputs

[boogheta]

• credits of an image (example : Légende
• 2 iframes inside an html block

[Yomguithereal]

First one is taken care of. The second one I don't understand.

[boogheta]

The second one was related to this blogpost https://medialab.sciencespo.fr/actu/les-enjeux-de-linformation-a-lere-numerique-vus-par-les-francais/
At the end the network and the legends are two different iframes and it was quite tricky in the backoffice to get those to display next to the other and was posing problems when either addind two successive html blocks or two iframes within the same html block.
I ended up using an embed box followed by an html block to do what we needed.
Not sure what to propose though, that would require some testing