Extract customized mask from QA band

I developed a python lib [pymasker](https://github.com/haoliangyu/pymasker) to extract customized mask from Landsat 8 QA band, like

``` python
from pymasker import landsatmasker
from pymasker import confidence

# load the QA band directly
masker = landsatmasker('LC80170302014272LGN00_BQA.TIF')

# load a numpy array that contains band data
masker = landsatmasker(bandarray)

# Get mask with cloud pixels (high confidence) and cirrus pixels (high confidence).
mask = masker.getmultimask(cloud = confidence.high, cirrus = confidence.high)

# Save the result if you want.
masker.savetif(mask, 'result.tif')
```

I wrote a detail guide in [my blog](http://haoliangyu.github.io/2015/01/19/Making-masks-with-Landsat-8-Quality-Assessment-band-using-Python/).

How do you think adding similar functionality to landsat-util?