Unicode name completions

Previously, we provided completion for a unicode name with its relative symbol.

With this, now IPython provides complete suggestions to unicode name symbols.

As on the PR, if user types ``\LAT<tab>``, IPython provides a list of

possible completions. In this case, it would be something like:

'LATIN CAPITAL LETTER A',

'LATIN CAPITAL LETTER B',

'LATIN CAPITAL LETTER C',

'LATIN CAPITAL LETTER D',

This help to type unicode character that do not have short latex aliases, and

have long unicode names. for example ``Ͱ``, ``\GREEK CAPITAL LETTER HETA``.

This feature was contributed by Luciana Marques `ghpull:`#11583`.

Make audio normalization optional

Added 'normalize' argument to `IPython.display.Audio`. This argument applies

when audio data is given as an array of samples. The default of `normalize=True`

preserves prior behavior of normalizing the audio to the maximum possible range.

Setting to `False` disables normalization.

Issues closed in 7.4

GitHub stats for 2019/02/18 - 2019/03/21 (tag: 7.3.0)

These lists are automatically generated, and may be incomplete or contain duplicates.

We closed 9 issues and merged 20 pull requests.

The full list can be seen `on GitHub <https://github.com/ipython/ipython/issues?q=milestone%3A7.3>`__

The following 23 authors contributed 69 commits.

* anatoly techtonik

* Benjamin Ragan-Kelley

* bnables

* Frédéric Chapoton

* Gabriel Potter

* Ian Bell

* Jake VanderPlas

* Jan S. (Milania1)

* Jesse Widner

* jsnydes

* Kyungdahm Yun

* Laurent Gautier

* Luciana da Costa Marques

* Matan Gover

* Matthias Bussonnier

* memeplex

* Mickaël Schoentgen

* Partha P. Mukherjee

* Philipp A

* Sanyam Agarwal

* Steve Nicholson

* Tony Fast

* Wouter Overmeire