If the e2e tests error, they leave all of the e2e state around including containers and networks etc. 
We should clean this up when the tests shuts down, even if it exits in error.

This should address last night's failure. We've taken the perspective
of "the load generator shouldn't cause tests to fail" in recent
days/weeks, and I think this is just a next step along that line. The
e2e tests shouldn't test performance. 

I included some comments indicating the ways that this isn't ideal (it
is perhaps not), and I think that if test networks could make
assertions about the required rate, that might be a cool future
improvement (and good, perhaps, for system benchmarking.)

I think the `Sync` check covers our primary use case, and perhaps we
can turn this back on in the future after some kind of event-system
rewrite, or RPC rewrite that will avoid the serverside timeout.

These are mostly the timeouts that I think we're still hitting in CI. 

At this point, the tests (on master) pass on my local machine (which is quite beefy) so I think this is just the first in (perhaps?) a sequence of changes that attempt to change timeouts and load patterns so that the tests pass in CI more reliably.

The previous implemention of hybrid set testing, which was entirely my
own creation, was a bit peculiar, and I think this probably clears thins up.

The previous implementation had far fewer legacy nodes in hybrid
networks, *and* also for some reason that I can't quite explain,
caused a test case to fail.

The 0.35 release cycle renamed the 'fastsync' functionality to 'blocksync'. This change brings the configuration parameters in line with that change. Namely, it updates the configuration file `[fastsync]` field to be `[blocksync]` and changes the command line flag and config file parameters `--fast-sync` and `fast-sync` to `--enable-block-sync` and `enable-block-sync` respectively.

Error messages were added to help users encountering these changes be able to quickly make the needed update to their files/scripts.

When using the old command line argument for fast-sync, the following is printed

```
./build/tendermint start --proxy-app=kvstore --consensus.create-empty-blocks=false --fast-sync=false
ERROR: invalid argument "false" for "--fast-sync" flag: --fast-sync has been deprecated, please use --enable-block-sync
```

When using one of the old config file parameters, the following is printed:
```
./build/tendermint start --proxy-app=kvstore --consensus.create-empty-blocks=false
ERROR: error in config file: a configuration parameter named 'fast-sync' was found in the configuration file. The 'fast-sync' parameter has been renamed to 'enable-block-sync', please update the 'fast-sync' field in your configuration file to 'enable-block-sync'
```

This changes the focus of the e2e suite, to (roughly) focus on
configurations that are more well used. Most production users of
tendermint run ABCI application in process and the GRPC/socket methods
cover the vast majority of the remaining use cases. 

Perhaps we should consider drop support unix domain sockets in a
future release, but I think in the mean time it's useful to have the
tests *mostly* focus on the primary use cases.

In retrospect my previous implementation of this node, could get
unlucky and never find the correct node. This method is more reliable.

When revwing #6807 I assumed that `probSetChoice` worked this way. 

I think that the coverage of various configuration options should
generally track what we expect the actual useage to be to focus the
most test coverage on the configurations that are the most prevelent.

This is just a configuration change to default to using the new stack
unless explicitly disabled (e.g. `UseLegacy`) this renames the
configuration value and makes the configuration logic more clear.

The legacy option is good to retain as a fallback if the new stack has
issues operationally, but we should make sure that most of the time
we're using the new stack.

In the transaction load generator, the e2e test harness previously distributed load randomly to hosts, which was a source of test non-determinism. This change distributes the load generation to the different nodes in the set in a round robin fashion, to produce more reliable results, but does not otherwise change the behavior of the test harness.

I realized after my last commit that my change made a following line of code a bit redundant.

(alternatively my last change was redunadnt to the existing code.)

I took this oppertunity to make some minor cleanups and logging changes to the node changes which I hope will make tests a bit more clear.

This is a mostly-automated fixup using Comby (https://comby.dev) to remove 
lexically-obvious redundant error checks. No functional changes are intended.

To reproduce the core change:

    # Collapse redundant error check conditionals
    % comby -in-place 'if err != nil {
       return err
    }
    return nil' 'return err' .go

    # Fold out unnecessary error temporaries
    % comby -in-place ':[spc~^\s*]err :[~:?]= :[any]
       return err' ':[spc]return :[any]' .go

Fixes #6479 and related cases.

* test/fuzz: add test to reproduce found fuzz errors

This change does two things:
1. It fixes the json fuzzer to account for receiving array results. Arrays are returned by the rpc server when the input data is an array.
2. Adds a `fuzz_test.go` file and corresponding `testdata` directory containing the failing test case.

This seems like a reasonable way to add and track previous crash issues in our fuzz test cases. The upcoming stdlib go fuzz tool does effectively this automatically.

Having looked at our network address parsing and connection code, it
really looks like we're not doing anything on top of what the standard
library is doing (both in terms using `net.ParseIP` and also
`net.Dial`,) and I don't think we need to run the tests 2x the number
of times just to run through different areas of the standard
library. I think most of our users are going to be using IPv4, and
would be down to fully remove this dimension as well, if we find it's
making noise, but for now I think it's fine.

This tweaks sleeps around pertubations, based on a theory that our
tests with "kill" pertubations restart the nodes fast enough the peers
haven't marked it down when it tries to reconnect. In my local test
runs, this clears out *most* of the test failures that I've seen,
except for one evidence-related test-harness problem (which should be
handled separately.)

* remove counter app

* remove unneeeded ci

* lint fix

* modify tx sizes

* cleanup docs

* Update abci/cmd/abci-cli/abci-cli.go

Co-authored-by: Callum Waters <cmwaters19@gmail.com>

* Update docs/app-dev/getting-started.md

Co-authored-by: Callum Waters <cmwaters19@gmail.com>

* Update docs/app-dev/getting-started.md

Co-authored-by: Callum Waters <cmwaters19@gmail.com>

* bring back comment

* migrate to kvstore and not persistent

* remove unused func

* test persistent

Co-authored-by: Callum Waters <cmwaters19@gmail.com>

## Description

Expose p2p functions for use in the sdk. 

These functions could also be copied over to the sdk. I dont have a preference of which is better.