fix: intermittent failure in dfx start

[ghost]

This fixes #971

The dfx start process was sleeping for 3 seconds and then checking the replica status once. Sometimes locally, but more often on CI, this was not long enough.

This PR changes the startup process to check the replica status once per second for up to 30 seconds before reporting an error.

The first commit in this PR added 200 CI tests that just ran the following:

dfx new
dfx start
dfx stop

Please see the failures in the CI build for the first commit in this PR, which failed 4 times:

• build log

The CI error in the second commit was due to formatting.

Notice that all subsequent commits, which incorporated the fix, did not exhibit any failures during dfx start.

Thanks to @hansl for pointing out the likely cause and suggesting an approach.

Intermittent errors looked similar to this:

# An error occured:
# AgentError(
#     ReqwestError(
#         reqwest::Error {
#             kind: Request,
#             url: "http://127.0.0.1:8000/api/v1/status",
#             source: hyper::Error(
#                 Connect,
#                 ConnectError(
#                     "tcp connect error",
#                     Os {
#                         code: 61,
#                         kind: ConnectionRefused,
#                         message: "Connection refused",
#                     },
#                 ),
#             ),
#         },
#     ),
# )