Use error/1,2 instead of exit/1

[erszcz]

This plays much nicer with ct_tty_hook.erl included in new ejabberd_tests. Instead of getting just the error reason, we also get the stacktrace showing exact location of the error.

I didn't test in the shell/Common Test results all possible/expected outputs these changes cause, but this is not a change in functionality and only affects the amount of information in error logs. If any readability problems or duplicated logs show up we can fix them on the go.

Basically, the difference boils down to this example:

exit/1:

> catch exit(asd).
{'EXIT',asd}

error/1:

> catch error(asd).
{'EXIT',{asd,[{erl_eval,do_apply,6,
                        [{file,"erl_eval.erl"},{line,573}]},
              {erl_eval,expr,5,[{file,"erl_eval.erl"},{line,357}]},
              {shell,exprs,7,[{file,"shell.erl"},{line,674}]},
              {shell,eval_exprs,7,[{file,"shell.erl"},{line,629}]},
              {shell,eval_loop,3,[{file,"shell.erl"},{line,614}]}]}}

error/2:

> catch error(asd, [123, qwe, zxc]).
{'EXIT',{asd,[{erl_eval,do_apply,
                        [123,qwe,zxc],
                        [{file,"erl_eval.erl"},{line,573}]},
              {erl_eval,expr,5,[{file,"erl_eval.erl"},{line,357}]},
              {shell,exprs,7,[{file,"shell.erl"},{line,674}]},
              {shell,eval_exprs,7,[{file,"shell.erl"},{line,629}]},
              {shell,eval_loop,3,[{file,"shell.erl"},{line,614}]}]}}

There are more places in Escalus code base where exit/1,2 is used, but I believe it's appropriate there as it is deliberately used to kill a process other than the current one.

[pzel]

Great code, man!
That said -- we should be adding small unit tests for escalus features, so regressions don't creep in.

Essentially, your pull request comment could be 'codified' into a test or two.

You reply:

• Yeah, whatever -> Simon merges
• Yeah, in a sec -> SImon merges happily with your test(s) added to the pull request

[erszcz]

Yeah, whatever ;p