Fix mass and parallax 2

[semaphoreP]

Ok, I think the second try starting fresh was a lot more successful.

• Re-implemented @sblunt's changes to system.py.
• The results posteriors are always the same size, no matter if mass/plx are fixed.
• Similarly, OFTI's samples are also fixed size. I think the slight increase in memory usage is negligible here.
• MCMC now samples only the non-fixed parameters, but will return posteriors that include the fixed parameters.
• Under this philosophy, it is up to the samplers to handle the fixed parameters gracefully (system.compute_model() requires mass and parallax as params as a result right now, since the system right now does not care about fixed parameters).
• Also, I fixed a bug where the Ensemble Sampler wasn't using the priors (was only calling the likelihood function).

Addresses #56

[coveralls]

Pull Request Test Coverage Report for Build 495

• 54 of 57 (94.74%) changed or added relevant lines in 2 files are covered.
• 5 unchanged lines in 4 files lost coverage.
• Overall coverage decreased (-1.0%) to 85.89%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
orbitize/sampler.py	47	50	94.0%

Files with Coverage Reduction	New Missed Lines	%
orbitize/results.py	1	80.13%
orbitize/sampler.py	1	96.34%
orbitize/priors.py	1	80.58%
orbitize/init.py	2	42.11%

Totals
Change from base Build 480:	-1.0%
Covered Lines:	627
Relevant Lines:	730

---

💛 - Coveralls

[coveralls]

Pull Request Test Coverage Report for Build 495

• 54 of 57 (94.74%) changed or added relevant lines in 2 files are covered.
• 5 unchanged lines in 4 files lost coverage.
• Overall coverage decreased (-1.0%) to 85.89%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
orbitize/sampler.py	47	50	94.0%

Files with Coverage Reduction	New Missed Lines	%
orbitize/results.py	1	80.13%
orbitize/sampler.py	1	96.34%
orbitize/priors.py	1	80.58%
orbitize/init.py	2	42.11%

Totals
Change from base Build 480:	-1.0%
Covered Lines:	627
Relevant Lines:	730

---

💛 - Coveralls

[sblunt]

Looks great, @semaphoreP. Thanks for working through all these changes.