STM32 Timer: add kernel clocks support

[gautierg-st]

This PR adds support the support of kernel source clocks for the STM32 timer driver.

For most STM32 (with the exception of U0), the kernel source clock of the timers is the AHB clock if APB divider is 1 or twice APB clock otherwise.

For most series/socs/timers, TIMPCLK is the only kernel clock available. This is represented by the use of the NO_SEL macro in the clock property of the timer dts node. This should not be modified.
But some timers of some series/socs can use another kernel clock. In that case, the clock property is defined with a macro TIMx_SEL(0), and it can be modified by user to select another source.

Some series also have a timer prescaler (timpre property). When enabled, the kernel source clock is AHB if APB divider is 1 or 2, or four times APB otherwise.

With the source clock frequency calculation now integrated into the clock driver, the calculations that were present in the counter and the PWM drivers are removed.

This PR apply changes on all STM32 series that currently have timer support (this excludes N6 and U3, support can be added later).

[GeorgeCGV]

Considering that there are checks #if defined(STM32_TIMER_PRESCALER) shouldn't it be wrapped with:

#if DT_NODE_HAS_PROP(DT_NODELABEL(rcc), timpre)
...
#endif

[gautierg-st]

Removed STM32_TIMER_PRESCALER and replaced by DT_NODE_HAS_PROP everywhere

[GeorgeCGV]

IMHO: it was cleaner with a separately defined prescaler.

However, it is clear how you want to use it. In that case, the DT_NODE_HAS_PROP is not the way, my bad.
The property is bool, and based on #90609 (comment) it is better to stick with:

#define STM32_TIMER_PRESCALER_ENABLED	DT_PROP(DT_NODELABEL(rcc), timpre)

and use IS_ENABLED where appropriate:

if (IS_ENABLED(STM32_TIMER_PRESCALER_ENABLED)) {
    ...
} else {
    ...
}

and value checks:

#if STM32_TIMER_PRESCALER_ENABLED
    ...
#endif

instead of #if defined(STM32_TIMER_PRESCALER), as it is always defined when DT_PROP is used.

[sonarqubecloud]

Quality Gate passed

Issues
10 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
9.6% Duplication on New Code

See analysis details on SonarQube Cloud