Implement new USB Serial JTAG reset strategy

[JurajSadel]

This is mostly experimental, so opening as a draft.

I've been playing with (cc cc #691) and I was able to reproduce the issue with espflash AND esptool as well, following the same steps as mentioned here
I looked into esptool and specifically to this commit. esptool not using this fix as default for SERIAL_JTAG, it can be used manually with:
--after reset_watchdog

The reasons (from esptool perspective) it's not on by default are:

1. The RTC WDT hack is not available on all devices (e.g., the H2 doesn't have it).
2. On Linux, it causes the port to re-enumerate, meaning that instead of /dev/ttyUSB0, a new /dev/ttyUSB1 appears. This can cause the monitor to disconnect and not reconnect automatically.
3. On the C6 and H2 USB-Serial/JTAG controller, there is a bug that causes the port to disappear after a system reset and not reappear until a power cycle is performed. The RTC WDT reset is a system reset, so this issue occurs. So they left it by default only for USB-OTG and is optional for SERIAL_JTAG. On C6 it is NOT SUPPORTTED at all in esptool.

[JurajSadel]

Should I add a P4 here as well?

[SergioGasquez]

A few comments that build on top of cleanups of #795

[jessebraham]

I think this still needs a fair bit of work, let me play around with it a bit tomorrow and I will try to give some more concrete feedback.

[jessebraham]

Still not 100% happy with this, but I think it's in much better shape now at least. Remaining concerns are fairly minor, can be handled at another time (or just ignored 😁).

[SergioGasquez]

LGTM! Thanks