Add ADC support STM32L4xx and STM32G4xx series MCUs

[Cipulot]

Description

These changes are based on @sigprof branch that he mentioned in a comment under #19922.

I've included his original changes that handle the dummy conversion to avoid ADC errors and also implemented the pinToMux() for both L4 and G4 series.

Implemented a check for the default ADC_SAMPLING_RATE, defaulting to ADC_SMPR_SMP_2P5 since it's the fastest available as per 'hal_adc_lld.h' in ChibiOS and datasheets for the L4 and G4.

I removed the debug print he included since it's not needed.

Tested with G431CBU, and confirmed working with pins of ADC1 and ADC2. Appreciate if others will test on some of the L4 MCUs supported in QMK, since I have no access to them as of now.

Change of the documentation about ADC support in the STM series will follow when this approved.

Types of Changes

• Core
• Bugfix
• New feature
• Enhancement/optimization
• Keyboard (addition or update)
• Keymap/layout/userspace (addition or update)
• Documentation

Issues Fixed or Closed by This PR

• Analog support for STM32G431 #19922

Checklist

• My code follows the code style of this project: C, Python
• I have read the PR Checklist document and have made the appropriate changes.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have added tests to cover my changes.
• I have tested the changes and verified that they work and don't break anything (as well as I can manage).

[sigprof]

Looks like this was based just on the first commit (8435456), and not the whole branch (master...sigprof:qmk_firmware:adc-stm32l4xx)? At least the RP2040 breakage seems to be still there.

[sigprof]

As for the sampling rate, the problem is that the fastest sampling also requires the lowest impedance of the signal source, and the required impedance is much lower than what AVR (or even RP2040) would require. E.g., if you look at the G431 datasheet:

At 2.5 cycles there is even no valid value for the source impedance if a “slow” channel is used (anything except ADCx_IN1…ADCx_IN5).

One typical usage of ADC is to connect a potentiometer-based joystick, and the typical resistance of those potentiometers is 10 kΩ; this obviously needs a much higher sampling time than 2.5 cycles, otherwise the readings are significantly off. So one option is to set a value which would be appropriate for such source as the default; but the problem is that it would also depend on the ADC clock frequency, and there does not seem to be a simple way to take that into account (in theory different ADC might also have different clock frequencies). Another option is just to set the fastest value (as was done for other STM32 chips, even though the default ADC clocks for them can be vastly different) and hope that any joystick users would understand that they need to change that value to something slower to get proper readings.

[Cipulot]

I was actually unsure about the sampling rate situation myself tbh. An idea that I had was to check if features like joystick etc are enabled, and if so to print a message in the terminal. Adding a note in the documentation once I update that is gonna be helpful too, both in the joystick section and in the ADC specs.

[Cipulot]

Also @sigprof did you test the support for oversampling for ADCv3? I saw that you had something in a branch of yours but I didn't include it in this PR. If by your testing that's good I'll amend this PR to include that too.

[Cipulot]

It seems like the CI Build fails are all using RP2040 ADC. Errors arise in the ADC_SAMPLING_Rate #ifdef.... will investigate in order to fix this.