Added the option to print system time rather than RTOS time in ESP_LOGx functions (IDFGH-1773)

[badcf00d]

I mainly wrote this because I wanted to syncronize logs from an ESP32 with logs from a cloud platform that were timestamped in UTC, and I think other people would probably find this option useful too.

Obviously this time will just start from 0 unless you set the system time (with an SNTP sync for example), but assuming you do set the time it makes the logs much more useful when trying to match up events happening on multiple platforms.

Example from the hello world program:

I (168) cpu_start: Chip Revision: 1
I (170) cpu_start: Starting scheduler on PRO CPU.
D (0) intr_alloc: Connected src 25 to int 2 (cpu 1)
I (0) cpu_start: Starting scheduler on APP CPU.
D (171) heap_init: New heap initialised at 0x3ffe0440
D (171) heap_init: New heap initialised at 0x3ffe4350
D (171) intr_alloc: Connected src 16 to int 12 (cpu 0)
Hello world!
D (00:00:00.011) efuse: coding scheme 0
D (00:00:00.012) efuse: In EFUSE_BLK0__DATA3_REG is used 1 bits starting with 15 bit
D (00:00:00.013) efuse: coding scheme 0
D (00:00:00.013) efuse: In EFUSE_BLK0__DATA5_REG is used 1 bits starting with 20 bit
This is ESP32 chip with 2 CPU cores, WiFi/BT/BLE, silicon revision 1, 4MB external flash
Restarting in 10 seconds...

[projectgus]

Hi @Frosticles,

Thanks for contributing this, it looks like a very useful change for some users.

I have a couple of minor requests but overall looks great.

[badcf00d]

I've added in the requests from @projectgus, given it a quick test my end and all seems well. Please do comment on any other changes you'd like but otherwise I'm happy to merge this for now.

[projectgus]

Thanks for the quick update @Frosticles and for being patient while I got back to you.

A couple of small things, one idea I didn't think of the first time around (sorry).

Also a very minor thing: the IDF code style is to put the opening brace of a statement on the same line as the statement, ie

 if ((timestamp > 0) || (i == 0)) {

[projectgus]

newlib locks actually lazy initialize, so you don't need this init check (the lock will be initialized the first time _lock_acquire() is called.)

[badcf00d]

Ah ok, didn't know that, I'll remove this bit

[projectgus]

Thanks again @Frosticles . I've squashed the commits and pushed them into our internal review & merge queue. This PR will be updated again once that process is complete.

[badcf00d]

Apologies for the late commit, not sure how I missed that but cppcheck spotted that I was accessing that buffer array out of bounds, just pushed a fix.

[projectgus]

@Frosticles Thanks, can't believe we both missed that! Have updated our internal branch, should be merged soon (has been a national holiday in China so things got a little delayed).

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[projectgus]

Oops, sorry this wasn't closed earlier @badcf00d . Changes were cherry-picked as
23e9224 .