pvsystem clean up

[wholmgren]

The most important parts of this PR are

• Changed some parameter names in pvsystem.sapm to follow the rules discussed in consistent variable names and capitalization #37 and the wiki. Would be good for people to look at this and make sure that they're happy with how it looks in practice.
• pvsystem.sapm documentation is more complete. Some module parameters were misnamed and we were missing some others.
• The pvsystem tutorial goes from terrible to very bad, but might help with Port Tutorials  #30 and Initial release to PyPi #15.

Less important changes:

• I added the "effective irradiance" Ee to the pvsystem.sapm output DataFrame. Not sure if this name follows the rules or not. It's a parameter that's very specific to the Sandia model, so that makes me think that it's better to leave as Ee.
• pvsystem.sapm can now take a dict for a module specification.
• By line, the majority of this PR is PEP8 improvements (except the tutorial of course), especially focused on wrapping long lines. Many PEP8 issues still exist and are beyond the scope of this PR.
• Added a note about a difference with the Matlab function. The difference has existed since the original python port.

[bmu]

I see one thing: According to the wiki systemdef should return surface_azimuth and surface_tilt instead of surfaz and surftilt.

[wholmgren]

Thanks. Fixed that and made some other minor changes too.

[robwandrews]

This looks good! Thanks for the cleanup