- cache received rfbridge codes in the internal list, allow to operate on it via the rpn operators
- add `<N> <proto> <code> rfb_match`, matching when we receive specified protocol + code string at least N times
- add `<proto> <code> <proto> <code> rfb_sequence`, checking if specified protocol + code pairs happen in sequence
- add `<TIME> <N> <proto> <code> rfb_match_wait` - similar to `rfb_match`, but waiting for at least `TIME` (ms) via oneshot runner
- add `<proto> <code> rfb_info`, pushes code's latest timestamp and it's counter on the stack
- add `<proto> <code> rfb_pop`, which removes the specified protocol + code from the internal cache
- fix MQTT skip setting making RPN variables absent on initial connection
- default to no skip when receiving MQTT
(small issue still stands with us having non-clean MQTT session, broker will persist variable subscriptions even after unsubscribe event)

ref.
https://github.com/Portisch/RF-Bridge-EFM8BB1/wiki/0xA2 (desc: Learning OK)
https://github.com/Portisch/RF-Bridge-EFM8BB1/wiki/0xA3 (desc: Learning failed)
were used, but the pages did not reflect the actual implementation:

af1bddb3d8/inc/uart.h
...
RF_CODE_LEARN_KO = 0xA2
RF_CODE_LEARN_OK = 0xA3
...

- no need in template when size set as constexpr
- remove size param completely, deduce size based on buffer

- queueing efm8bb1 msg, pass around array as reference since the size is fixed
- make use of learn timestamp

- for some reason we were never checking it's output
no need to increase +1 since we write the exact number
- allow hexEncode output be decoded, enforce even-ness
- raw msg must have at least 3 bytes - start+code+end

- RF_... -> RFB_...
- rework rcswitch integration, support variable length payload
- rework rfbridge parser (not tested)
- rework settings scanning routine when trying to match rf payload
  with the relay ID
- update build tests

resolve #2326

As this is no longer an option since PIO 4.4.0
ref.
04694b4126
https://docs.platformio.org/en/latest/core/history.html

*...I have known you for so long
When we ran in circles...*

Remember MAC for the current IP sequence number
(e.g. lease 0 -> xxx.xxx.xxx.2, lease 1 -> xxx.xxx.xxx.3 and etc.)

~500 bytes of code

As codacy points out, move stype and switchType to the LIGHT_PROVIDER != NONE block

- reduce overall size of the structure, store a single required entity
  inside of Api object itself.
- tweak internal functions to expect a certain path size
- (kind of a hack) add manual calls to vector<Api>::reserve() as we go over
  the current capacity and vector increases it's size times 2.
  e.g. for 9 relays, we would allocate space for 32 Api objects
  when vector size goes from 16 to 32, after we add 18 Api objects with relay + pulse
- (breaking) json calls on a separate path, don't waste time encoding
  a single entity as json object when we can encode more things

- no need for smartptr when instance is static
- default ctor, set default values in the body

Co-authored-by: mkozlowski <gitcmt@empeka.pl>

* rpn: bump to rpnlib 0.24.1

- `i` and `u` suffixes for numbers in expressions, parse as Integer or Unsigned respectively
- `checkedTo...()` method variants for numeric conversions
- strings can contain escape sequences (\x61\x62\x63, \n, \t, \r)
- improve float number parsing
- more consistent whitespace checks, tokens also can be separated by \n or \t
- various parser fixes

* fix captures

## [0.23.0] 2020-07-26
### Added
- `p` operator to print the top of the stack via debug function
- `&var` syntax to create variable reference in expression
- `=` operator for variable assignment in expression
- `exists` operator to check for variable existance (only for references)
- `deref` operator to convert variable reference into a value (only for references)
- Allow to use either float or double as floating type, parse numbers in expressions as specified type
- Add boolean type, parse `true` and `false` in expressions
- Add null type, parse `null` in expressions
- Add string type, parse double-quoted `"string"` in expressions
- Add integer and unsigned integer type, used in operators
- Allow to configure underlying types from rpnlib\_config.h and -D... flags
- Return `rpn_error` from operators, split error types into categories
- Create a new stack by using `[` keyword. Move stack contents into the previous stack + size by using `]`.

### Changed
- Stack structure no longer holds raw `float`, but internal `rpn_value` type
- rpn\_... setter and getter methods use `rpn_value` type
- Operator functions return `rpn_error` type, allowing to return both value and operator errors
- Variables in expressions are no longer required to exist when using `&var`
  Expression will automatically create the variable, set it to `null` and push it's reference on the stack
- It is possible to create 'reference' stack values
- Improve precision of `e` and `pi`

### Fixed
- Proper value for `e` constant
- Allow to use multiple contexts simultaniously, replace `rpn_error` and `rpn_debug_callback`
  with the current `rpn_context` members `error` and `debug_callback` respectively

Add support for BME680 using libalgobsec proprietary algorithms for precise Indoor Air Quality (IAQ) measurement. Unlike traditional CO2 sensors - and good ones are expensive - it measures nearly all VOCs compounds in the air (plus other gases) and compensates those measurements with its built-in temperature and humidity sensors to determine indoor air quality.

Co-authored-by: Max Prokhorov <prokhorov.max@outlook.com>

Using generic size type instead of unsigned char

This reverts commit d57eb0721b.

We can use raw value as-is. While the lack of unit is true for the value represented as cos θ , we don't specifically have any other sensors using it as such. As comment suggest (huh), we want to express the thing as a percentage.