Rcc improvement

[AndreySmirnov81]

No description provided.

[AndreySmirnov81]

Sorry.
@burrbull, please explain why "impairment"?

[burrbull]

Because I don't see any "improvement".

1. If you delete RCC dependency, you need mark enable & reset with unsafe
2. What sense of &self requirement? (not sure about this)
3. steal(). Again.

[AndreySmirnov81]

@burrbull

1. What is the reason unsafe?
   I think that after switching to using bit-banding, the need for &rcc disappeared.
2. &self is needed so that only the owner of the corresponding peripheral can do enable and reset for it.
   This will help to avoid mistakes like this:

   pub fn initialize(
      ...
      tim2: pac::TIM2,
      ...
   ) -> Self {
      ...
      let rcc = unsafe { &(*pac::RCC::PTR) };
      pac::TIM1::enable(rcc);
      pac::TIM1::reset(rcc);
      ...
   }

3. A temporary way out of the situation. In the future, it is necessary to make a similar change in the stm32-usbd create: Usb Peripheral::enable() -> UsbPeripheral::enable(&self).
   Is steal() acceptable here https://github.com/stm32-rs/stm32f3xx-hal/blob/565b43ceab7b8255c122e82f4f1f1f2b3e85f11b/src/serial.rs#L1270 ?
   I believe that using steal() in certain situations is quite acceptable.

[AndreySmirnov81]

Please review and merge.

[AndreySmirnov81]

@burrbull
Do you agree with the changes?

[AndreySmirnov81]

@therealprof, @TheZoq2
Please review