Refactor all frequencies from MHz (double) to Hz (uint32_t)

[non-det-alle]

In the LoRaWAN MAC protocol specifications, in the LoRa devices PHY hardware abstraction layer libraries, in the gateways' packet forwarder protocol code and in the network server configurations and internal representation, frequency values are encoded using 4 bytes (and in Hz). However, the module currently uses MHz values in floating point form to represent frequencies.

Aligning with real implementations slightly simplifies encoding and prevent floating point arithmetic issues. From personal experience, I once got a bug because multiplying frequency * 0.01 instead of frequency / 100 returned a .99999999-like value, and as a consequence all comparisons to retrieve the correct LogicalLoraChannel object yielded false.

This PR converts the type of frequency values used throughout the module from double to uint32_t, and it appends Hz to names of class members and function signature parameters representing frequencies (i.e. ending up in the API documentation).

This PR is an effort to break #135 into more digestible pieces.

[codecov]

Codecov Report

Attention: Patch coverage is 76.92308% with 39 lines in your changes missing coverage. Please review.

Project coverage is 77.66%. Comparing base (82ea58a) to head (5567991).

Files with missing lines	Patch %	Lines
model/mac-command.cc	0.00%	11 Missing ⚠️
model/end-device-lorawan-mac.cc	27.27%	8 Missing ⚠️
model/gateway-lora-phy.cc	50.00%	4 Missing ⚠️
model/logical-lora-channel-helper.cc	60.00%	4 Missing ⚠️
model/lora-frame-header.cc	0.00%	4 Missing ⚠️
model/class-a-end-device-lorawan-mac.cc	57.14%	3 Missing ⚠️
model/end-device-status.cc	71.42%	2 Missing ⚠️
model/logical-lora-channel.cc	88.88%	1 Missing ⚠️
model/lora-channel.cc	66.66%	1 Missing ⚠️
model/sub-band.cc	87.50%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff            @@
##           develop     #179   +/-   ##
========================================
  Coverage    77.66%   77.66%           
========================================
  Files           66       66           
  Lines         6831     6833    +2     
========================================
+ Hits          5305     5307    +2     
  Misses        1526     1526           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[non-det-alle]

Codecov fails because the test coverage of changed lines is subpar to the overall coverage of the module. I'll try to add some tests to fix this while we are at it.

[pagmatt]

a few variables seem to be still double

[non-det-alle]

Thank you, I must have missed them. I'll check everything again.

[non-det-alle]

@pagmatt if you agree, before merging this I'd like to both merge #180 and the tests required to cover the lines indicated by codecov (I have yet to create a PR for this). It feels the right order of actions given the dependency of changes bug<-tests<-this. Afterwards I'll take care of merging changes.