improve speed of Location.get_airmass

[wholmgren]

• Closes slow performance when assigning to empty DataFrame in Location.get_airmass and ModelChain.prepare_inputs #502
• I am familiar with the contributing guidelines.
• Fully tested. Added and/or modified tests to ensure correct behavior for all reasonable inputs. Tests (usually) must pass on the TravisCI and Appveyor testing services.
• Updates entries to docs/sphinx/source/api.rst for API changes.
• Adds description and name entries in the appropriate docs/sphinx/source/whatsnew file for all changes.
• Code quality and style is sufficient. Passes git diff upstream/master -u -- "*.py" | flake8 --diff
• New code is fully documented. Includes sphinx/numpydoc compliant docstrings and comments in the code where necessary.
• Pull request is nearly complete and ready for detailed review.

[mikofski]

I think an even faster way to do this would be:

airmass = pd.DataFrame(
    {'airmass_relative': airmass_relative,
     'airmass_absolute': airmass_absolute},
    index=solar_position.index)

Nope I was wrong, it's already faster the way it is, and it keeps the correct order for all pythons b/c py2 dictionaries are unordered

[wholmgren]

@mikofski see the linked issue for timings. I slightly prefer this approach for style reasons. Happy to change if most people prefer the other approach.

[mikofski]

Okay I was wrong, I like it the way it is now. Thanks Will!