py-h5py: updated to 2.9.0

What's new in h5py 2.9

New features
* A convenient high-level API for creating virtual datasets, HDF5 objects which
  act as a view over one or more real datasets.
  See :ref:vds for details.
* :class:File can now be constructed with a Python file-like object,
  making it easy to create an HDF5 file in memory using :class:io.BytesIO
  See :ref:file_fileobj for details.
* :class:File now accepts parameters to control the chunk cache.
  See :ref:file_cache for details.
* New options to record the order of insertion for attributes and group entries.
  Iterating over these collections now follows insertion order if it was
  recorded, or name order if not.
* A new method :meth:Group.create_dataset_like to create a new dataset with
  similar properties to an existing one.
* Datasets can now be created with storage backed by external non-HDF5 files
* Lists or tuples of unicode strings can now be stored as HDF5 attributes
* Inspecting the view returned by .keys() now shows the key names, for
  convenient interactive use.

Exposing HDF5 functions
* H5LTopen_file_image as :func:h5py.h5f.open_file_image.
* External dataset storage functions H5Pset_external, H5Pget_external
  and H5Pget_external_count as methods on :class:h5py.h5p.PropDCID

Bugfixes
* Fix reading/writing of float128 data.
* Converting data to float16 when creating a dataset.

Support for old Python
Support for Python 3.3 has been dropped.
Support for Python 2.6 has been dropped.

(adam)