refactor(core): simplify fonts handling

[obrusvit]

This PR aims to further simplify the font handling in core.

The main objective was to get rid of the numerous macros, which are quite cumbersome to work with. This is achieved by introducing struct font_info_t which is filled with needed data for each font during font generation by gen_font.py. Then there is a function get_font_info defined In the core/embed/lib/fonts/fonts.c which is used to access font data.

Unfortunately, I was unable to get rid of the FONT_MAX_HEIGHT calculation as it must be done in compile time. However, there is only one usage so getting rid of it is perhaps feasible.

trezor-firmware/core/embed/lib/buffers.h

Line 32 in 02533aa

#if TEXT_BUFFER_HEIGHT < FONT_MAX_HEIGHT

TODO:

• _UPPER fonts seems not effective
• problems with translated strings

[github-actions]

core UI changes	device test	click test	persistence test
T2T1 Model T	test(screens) main(screens)	test(screens) main(screens)	test(screens) main(screens)
T3B1 Safe 3	test(screens) main(screens)	test(screens) main(screens)	2724
T3T1	test(screens) main(screens)	test(screens) main(screens)	test(screens) main(screens)
All	main(screens)

[TychoVrahe]

i think we will drop entire buffers.c/h when we get rid of old rendering, which will hopefully be soon enough.

[obrusvit]

All UI-diffs on TS3 seems to be caused by now usage of BOLD font in Size: x bytes which is actually correct.

[obrusvit]

Can the change of order in translation jsons cause some problems moving forward @matejcik ? If so, I can revert it and just define the enum font_id_t starting from 1 and not change the order. I simply though that the new ordering is a little bit more logical, e.g. having NORMAL and NORMAL_UPPER in succession.

The reason to start from 0 is to use the last enum variant FONTS_COUNT.

[matejcik]

yeah the ids are baked into the translation blobs, this would really mess up the upgrade/downgrade scenario. please revert.

aside, I believe that there was a reason for skipping 0 for font index, but perhaps it doesn't apply anymore with this refactor?

[obrusvit]

I reverted the re-assignment of font IDs. The PR is ready for review.

For future reference: after we drop buffers.h/.c when getting rid of "old rendering", then we can completely remove these definitions for all fonts (also remove this from gen_font.py and also remove the FONT_MAX_HEIGHT calculation in fonts.h
#define Font_PixelOperatorMono_Regular_8_MAX_HEIGHT 8