GitLab

Response from wallet after API call looks like this:

{
  "id": 0,
  "result": [
    <some interesting information>
  ]
}

In the vast majority of cases TestTools users needs only <some interesting
information>. So tests contains a lot of boilerplate code:

result0 = wallet.api.some_call()['result']
result1 = wallet.api.some_other_call()['result']

Everytime tester needs to fetch 'result' field. This commit changes default
behavior to return only 'result' field. So above code will be simplified to
following form:

result0 = wallet.api.some_call()
result1 = wallet.api.some_other_call()

If someone will need to read 'id' field, they can do:
id_ = wallet.api.some_call(only_result=False)['id']

When test was parametrized, every time it was run in the same directory.
It causes problems when test were run simultanously. This change makes,
that every test (also parametrized) creates its own directory.

E.g. replace "-d" with "--daemon".