Prevent blocking when Infineat watchface is charging

[SteveAmor]

Fixes #2245
Blocking removed for the edge case where it reverts to the clock on wakeup if no app is loaded and the watchface is Infineat and the watch is charging. In this particular case, there is a small UX watchface transition glitch on wakeup. The alternative is blocking code which often results in the watch rebooting.

[github-actions]

Build size and comparison to main:

Section	Size	Difference
text	373040B	16B
data	948B	0B
bss	22536B	0B

Run in InfiniEmu

[mark9064]

Thanks for the patch! If we don't find a better solution before 1.16, I think this is a good temporary measure
It sounds like there are two core issues though:

• The Infineat charging animation is capable of using 100% drawing bandwidth (CPU/IO), and therefore causes LVGL to never finish. This sounds suspicious and should be reworked. Drawing a battery indicator should not be so resource demanding
• The watchface load loop is overoptimistic and locks up DisplayApp when faced with 100% load

Both issues should be fixed. I've just looked on InfiniEmu and indeed the CPU is in the Running state the entire time when charging on Infineat (though my laptop CPU runs it very slowly, so this may be a factor), so this issue is pretty much confirmed. The watchface load loop should be fixed in LVGL. I have some testing code for this but it's super ugly. More broadly I need to really understand the LVGL render pipeline so I can point to the exact parts we need to change (i.e guaranteeing the watchface Refresh() runs exactly once, and then also that drawing this one refresh is fully finished)

[mark9064]

Closing as #2256 is merged now :)