This does not constitute any kind of stable API promise.
But it might allow people to use our arti client code in novel ways.

This is just code motion and the minimal fixups.

Make ArtiConfigBuilder contain a TorClientConfigBuilder

See merge request tpo/core/arti!401

Follow-up from arti!318

See merge request tpo/core/arti!402

Make update_config only conditionally exported; add semver-status update.

Add flatfile DirMgr

See merge request tpo/core/arti!318

This is an API break: now one must use `.tor()` to access the Tor
configuration parts.

But it is not a config file format break, because `#[serde(flatten)]`.

As per #371

tests/chutney: Merge chutney-launching scripts

See merge request tpo/core/arti!396

It turns out that the clone-and-hack in tests/chutney/arti-bench
was not identical to the original in tests/chutney/setup.

I diffed the two and preferred the version from arti-bench.

Formerly, the "launch a chutney network" code was duplicated in
"setup" and "arti-bench", since "setup" always launched an arti
proxy, while "arti-bench" didn't want that functionality.

Now the "setup" script launches an arti proxy conditionally,
depending on whether the "proxy" argument is given.

Drop config back-conversion from FooConfig to FooConfigBuilder

See merge request tpo/core/arti!391

arti-testing: support for simulated TCP breakage

See merge request tpo/core/arti!392

arti-bench: don't allocate a separate receive-buffer for each job

Closes #391

See merge request tpo/core/arti!395

tor-chanmgr: Do not allocate an Internal error unless we mean it.

Closes #383

See merge request tpo/core/arti!394

This makes heap profiling more viable.

Closes #391.

Previously we'd allocate an error as a place-holder here, but it's
not a great idea to do that with a `Bug`: each `Bug` stores a whole
stack trace, which uses a whole pile of allocations to construct.

Now we keep an `Option<Error>` instead.

Found while heap profiling.

Closes #383.

Note the dependency on PCRE support in Git implied by the use of grep -P in the git hooks

See merge request tpo/core/arti!393

Add missing dependencies to CONTRIBUTING.md

See merge request tpo/core/arti!389

Fix expand tilde and invalid path chars on windows

See merge request tpo/core/arti!274

Add instructions for hooking up Tor Browser to arti on Windows

See merge request tpo/core/arti!388

Replace the recapitulation of TorClientConfig fields in ArtiConfig and
instead just have it contain one.  This is part of #374.

The conversions from ArtiConfig back to ArtiConfigBuilder and
TorClientConfigBuilder would need to change, but, since we don't want
them anyway,

No longer impl Deserialize for ArtiConfig.  (As per #371 this will
want to become a private type.)

No longer impl From<ArtiConfig> for ArtiConfigBuilder and
TorClientConfigBuilder.  And abolish tests of that code.

(This all has to be in one commit, because previously
ArtiConfig::tor_client_config used the validated-to-builder config
retcon.)

This is needed according to #372, where we observe that builders ought
not to be generated from validated structs.  So we need this
conversion.

I used
  git-grep -P '\#\[serde\((?!default|deny_unknown)'
to find places where I needed to add additional attributes on the
builder method fields.

This is currently a bit duplicative, but when #371 is completely done,
the validated (non-builder) configs won't need to be Deserialize any
more.

This is part of #371 and #372.

We are going to want to specify custom attributes on fields of the
builder struct.  This feature was missing from derive_builder.

This commitid is the current head of my MR branch
  https://github.com/colin-kiegel/rust-derive-builder/pull/237
  https://github.com/ijackson/rust-derive-builder/tree/builder-field-attrs
Using the commitid prevents surprises if that branch is updated.

We will require this newer version of derive_builder.  The version
will need to be bumped again later, assuming the upstream MR is merged
and upstream do a release containing the needed changes.