Complete firmware refactoring

[marshalab]

I had errors compiling the firmware for esp32. I had to rework several functions ledWrite. Then I could not understand why the indicators did not work, I rewrote them completely. As a result, I completely combined the firmware. If this is useful to someone, I will be glad to share. But unfortunately I did not conduct a full test, there is little time to assemble the machines and there are no machines themselves. Perhaps I am doing something wrong to publish the code, I am just learning to use github.

[thias15]

Thank you for sharing this — and welcome to GitHub! The amount of work here is significant and the ESP32 ledcWrite rework sounds genuinely useful.

A few things would need to happen before this could be merged, though:

1. The PR documents several broken features in README.md (indicators, bumper, ultrasonic-to-phone, status query). Those are core firmware features for the OpenBot app — could you investigate and fix those before we look at merging? Without them working, replacing the current firmware would be a regression for everyone running it today.
2. End-to-end test on a real device. You mention you couldn't test fully. We'd really need someone to verify the refactored firmware on at least a couple of variants (e.g. DIY-Nano and an ESP32-based RTR/DIY) before swapping out the canonical sketch. Is anyone available to help test, or could the OpenBot team coordinate a test once the known glitches are fixed?
3. README.md shouldn't contain Russian-language dev notes. The "Известные глюки" / "Проверено - работает" sections look like local notes — they should be removed (or the info promoted into the PR description / a GitHub issue) before merge.
4. The new firmware/README.en-EN.md is redundant. This repo's translation convention is that README.md is English and README.<locale>.md are translations (e.g. README.zh-CN.md, README.de-DE.md). There's no en-EN file in the rest of the project — please drop this one.
5. There are merge conflicts with current master. Would need a rebase.
6. Tip: rather than pushing PRs from your master branch, create a topic branch (e.g. firmware/esp32-refactor) — that way you can keep master in sync with upstream and iterate on the PR independently.

Happy to keep this open while you address the broken features — your ledcWrite fixes for newer ESP32 cores will be valuable to land once the rest is stable.