Tidying up calc.py

[jolyonb]

This does a bunch of tidying in calc.py.

• evaluator now has default arguments for variables, functions and suffixes.
• Made all negative signs limited to just one, and made a bunch of tests to ensure compliance.
• Moved one-line pyparsing actions to lambda functions.
• Cleaned up dump_parse_result
• Gave names to all symbols in the parser tree, so that you don't see any more ITEM tags in the debug dump. Eg., 1+2^-2*sin(x) now parses to the following debug tree:

<sum>
  <number>
    <num>1</num>
  </number>
  <op>+</op>
  <product>
    <power>
      <number>
        <num>2</num>
      </number>
      <op>-</op>
      <number>
        <num>2</num>
      </number>
    </power>
    <op>*</op>
    <function>
      <funcname>sin</funcname>
      <arguments>
        <variable>
          <varname>x</varname>
        </variable>
      </arguments>
    </function>
  </product>
</sum>

I was trying to get the variable tags to work as <variable>x</variable>, but I couldn't seem to get that to happen...

[coveralls]

Coverage remained the same at 100.0% when pulling e56a40d7919691a50658e9e87d0d224644026ad0 on calc_tidy into c358c71ee79765250a164dc2e3bdc734cfb181df on master.

[jolyonb]

@ChristopherChudzicki This is ready for proofing now.