WIP removed the apparently-unnecessary anyOf blocks

[gmabey]

Note that (for clarity on the substantive edits) this does not include appropriate indent-level changes.
Closes #337

[Teque5]

I think you should include the indent changes here too. I usually do a round-trip in python but I think people also use jq for this.

[Teque5]

I don't think you meant to change these. This number is 2**63 + 192. Was previously max signed integer.

[gmabey]

:-O blames jq
At least the mystery is solved regarding how the previous value for maximum came to be -- it was surely int64_t(double(2**64-1))

[gmabey]

I suppose if we were smart we would use an integer small enough such that int64_t(double(maximum_value)) < (2**63-1) ...

[gmabey]

I now notice that the jq-ification rewrote 1e12 -> 1000000000000 which I dislike

[gmabey]

Looking for feedback from @777arc (or anyone who can give these changes a spin).
Also, I'm thinking harder about that joke/threat I made to suggest a number less than 2**63-1 in an effort to sidestep this issue in the future (jq-ification). Experimentally I found it to be 9223372036854774784 (tentatively).

[777arc]

FYI you can see the rendered docs by going to the "Build docs" check then clicking Summary on the left, then click "compiled_docs", and open sigmf-spec.html. The core:sample_rate renders the same, everything seems to render the same.