Update the documentation

[CyrilleB79]

I have asked to have Speech History add-on included in translation framework (cf. jscholes/nvda-speech-history#23).

I realize that the add-ons need to be based on the add-on template to be compatible with the translation system. But it is not documented in the readme of this repo. More specifically:

• The files sconstruct and buildVars.py are required since they are checked in 01_common.sh.
• The command scons pot mergePot should be runable and successful (run in 01_addon2svn.sh)
• Is there a need/a check that scons generate a .nvda-addon file?

Also the readme does not indicate which file is used to generate the work for translators. There is the .po file generated by scons pot mergepot. But there is also the English webpage file: srt/website/addons/addonname.mdwn.

@jscholes, if you agree to work on jscholes/nvda-speech-history#23, you may write your feedback here if something else is not documented.

Cc @nvdaes, @feerrenrut