docs: core-geth features, generated API method docs

[meowsbits]

Working on adding some documentation about what makes core-geth core-geth.

Resolves #308

[meowsbits]

Nice to have:

• https://squidfunk.github.io/mkdocs-material/reference/content-tabs/
• eth: added trace_call to trace on top of arbitrary blocks ethereum/go-ethereum#21338 (see MethodInvocation section, is nice)

[meowsbits]

I'm gonna squash this and mark ready for review.

[meowsbits]

Aside from the generated API method docs (./docs/JSON-RPC-API/modules/*.md), here's the most significant content addition: https://github.com/etclabscore/core-geth/blob/docs/core-geth-features/docs/core/index.md.

tree for reference:

> tree docs
docs
├── audits
│   ├── 2017-04-25_Geth-audit_Truesec.pdf
│   ├── 2018-09-14_Clef-audit_NCC.pdf
│   ├── 2019-10-15_Discv5_audit_LeastAuthority.pdf
│   └── 2020-01-24_DiscV5_audit_Cure53.pdf
├── core
│   ├── alltools.md
│   ├── evmc.md
│   └── index.md
├── developers
│   ├── build-from-source.md
│   ├── create-new-release.md
│   ├── documentation.md
│   └── versioning.md
├── getting-started
│   ├── installation.md
│   └── run-cli.md
├── img
│   └── favicon.png
├── index.md
├── JSON-RPC-API
│   ├── index.md
│   ├── modules
│   │   ├── admin.md
│   │   ├── debug.md
│   │   ├── ethash.md
│   │   ├── eth.md
│   │   ├── miner.md
│   │   ├── net.md
│   │   ├── personal.md
│   │   ├── trace.md
│   │   ├── txpool.md
│   │   └── web3.md
│   └── openrpc.md
└── tutorials
    └── private-network.md

8 directories, 28 files

[meowsbits]

Just putting this out there... the generated docs in docs/JSON-RPC-APIs/modules/ aren't deterministic. And this bothers me. It's going to be practically impossible to review diffs, for example. The nondeterminism is caused, I believe exclusively, by Go's non-guaranteed map ordering.

[ziogaschr]

Would be nice if they can be deterministic, though I can leave with that as this PR brings a lot of value in the docs.

[meowsbits]

cc5afb0 begins the ordering solve. In fact, so far, it seems to be 100% effective. I don't think its a complete solution because the template also includes information (as Raw JSON) from the plain old json.MarshalIndent which shouldn't guarantee ordering, but at least for me, for now, that doesn't seem to be an issue. Note this commit's diff with de73918, showing that only the generated-at timestamps and expected lines are modified.

Would be nice if they can be deterministic, though I can leave with that as this PR brings a lot of value in the docs.
#306 (comment)

With these last two commits, I'm happy merging this and following up later if the json.MarshalIndent causes undesirable noise. For now, close enough.

[meowsbits]

@ziogaschr PTAL #306 (comment). I'm OK with merging if you are.

[ziogaschr]

Looks fine for me at the moment. Nice work man. We can do a new PR when and if needed.