Add support for LT8722

[rccalam]

PR Description

• This adds driver support and device tree bindings for LT8722 full bridge DC/DC converter.
• Datasheet: LT8722
• Tested on RPI4 using the following demo board: DC3145A

PR Type

• Bug fix (a change that fixes an issue)
• New feature (a change that adds new functionality)
• Breaking change (a change that affects other repos or cause CIs to fail)

PR Checklist

• I have conducted a self-review of my own code changes
• I have tested the changes on the relevant hardware
• I have updated the documentation outside this repo accordingly (if there is the case)

[nunojsa]

here it goes my first round

[nunojsa]

instead of open coding FIELD_PREP(). Can we make sure val is already properly set (in the same way we would pass it to regmap_update_bits()) and then have a one liner:

reg_val = (reg_val & ~mask) | (reg_val & mask);

[rccalam]

V2:

DT Bindings

• Renamed en-gpios to enable-gpios and swen-gpios switch-enable-gpios for clarity
• Improved descriptions for enable-gpios and switch-enable-gpios to reflect their specific roles
• Ensured description is the first attribute in property definitions
• Refactored description remove default value mentions, replacing them with actual default definitions in the schema
• Updated adi,switch-frequency and adi,duty-cycle-range to use descriptive string values for clarity

Kconfig

• Updated Kconfig to reflect that LT8722 depends on both SPI and OF

Driver

• Sorted header files alphabetically
• Adopted correct kernel style single line comment
• Removed extra lines
• Removed redundant GPIO direction setting
• Replaced ret == 0 with !ret
• Moved device tree properties retrieval into a separate helper function.
• Used fsleep()
• Used gpiod_set_value_cansleep()
• Introduced LT8722_RAMP_STEP constant to replace hardcoded value
• Removed unnecessary NULL check for SPI device
• Ensured all declarations are positioned before any code
• Deleted _lt8722_crc helper function
• Prepared val before passing value to lt8722_reg_write_mask
• Removed use of clamp()
• Removed SPI setup

[nunojsa]

Looks better. take my inputs and send it upstream... I think it's already good enough to be sent. Of course you'll get more feedback from the maintainers.