Fix LoRaWAN 1.1 support

[rvolosatovs]

Summary

References #2171

Changes

• Fix 1.1 uplink MIC calculation naming (aligned it with the spec directly) - block called b0 currently is actually b1 according to the spec, that also caused the confusion in making me think that the implementation is incorrect, let's fix this to avoid this happening in the future.
• Fix DeviceModeInd being marked as "initiated by NS"
• Add myself to pkg/crypto CODEOWNERS

Checklist

• Scope: The referenced issue is addressed, there are no unrelated changes.
• Compatibility: The changes are backwards compatible with existing API, database and configuration, according to the stability commitments in README.md.
• Testing: The changes are covered with unit tests. The changes are tested manually as well.
• Documentation: Relevant documentation is added or updated.
• Changelog: Significant features, behavior changes, deprecations and fixes are added to CHANGELOG.md.
• Commits: Commit messages follow guidelines in CONTRIBUTING.md, there are no fixup commits left.

[rvolosatovs]

Tested this with FF1705 running mbed-os with patch from ARMmbed/mbed-os#12762 applied on top of feature-lorawan-1-1

[coveralls]

Coverage decreased (-0.0006%) to 73.528% when pulling cb57888 on rvolosatovs:fix/lorawan-1.1 into 0ec93de on TheThingsNetwork:master.

[htdvisser]

The changes look good to me, but the lack of unit tests is a bit "smelly" IMO.

[rvolosatovs]

The InitiatedByDevice fields in pkg/encoding/lorawan/mac.go is not something we can or should unit test, since that's just static data. This will be tested in Network Server indirectly though in #2221

Regarding pkg/crypto I agree, but I'd expect the original implementation to be already covered by existing unit tests? I'll see if I can add something quick, it would be great to have some single "source of truth" to test against, unfortunately, the spec does not contain that, so the best we can do is to just run the function once and then test that it gives the same output in the test given same input, so we test against the result of the function we're testing, which is not ideal. Since there are no actual functional changes in pkg/crypto, no reason to add tests here