The plan goes like this:
- eeschema still uses int in decidegrees
- all the other things internally use double in decidegrees (or radians
  in temporaries)
- in pcbnew UI the unit is *still* int in decidegrees

The idea is to have better precision everywhere while keeping the user with int i
angles. Hopefully, if a fractional angle doesn't come in from the outside, everything
should *look* like an integer angle (unless I forgot something and it broke)

When the time comes, simply updating the UI for allowing doubles from the user should
be enough to get arbitrary angles in pcbnew.

New conversion routines and sin/cos implementation for angles in decidegrees

- Removed spurious int casts (these are truncated anyway and will break
  doubles)

- Applied the Distance, GetLineLength, EuclideanNorm, DEG2RAD, RAD2DEG
  ArcTangente and NORMALIZE* functions where possible

- ArcTangente now returns double and handles the 0,0 case like atan2, so
  it's no longer necessary to check for it before calling

- Small functions in trigo moved as inline

Charras beat me on fixing the include/sch_item_struct.h include issue from 4069

In particular the new mechanism for handling extended color palettes is in place,
included renaming the ini keys and saving the color name instead of its index; this means better forward compatibility with palette changes.

Since ini keys are changed, colors will be reset

Drc_On => g_Drc_On
Route_Layer_TOP => g_Route_Layer_TOP
Route_Layer_BOTTOM => g_Route_Layer_BOTTOM
Track_45_Only_Allowed => g_Track_45_Only_Allowed

Pcbnew: fix Bug #1148785 (pcbnew crashes when using only one layer in autorouter ) , and a minor bug in dialog cleanup.

Code cleaning in autoroute functions.

autoroute functions: code cleaning

Some code cleaning in autoroute functions.

Minor code cleaning in autoroute files.

fix nanometer compatibility.
minor code and comments cleaning (although theses files need a lot of code cleaning)

* EDA_RECT and EDA_ITEM completely encapsulated.
* Removed unnecessary EDA_ITEM member m_Selected and replaced functionality
  with SELECTED flag bit.

  * Removed Pcb_Frame argument from BOARD() constructor, since it precludes
    having a BOARD being edited by more than one editor, it was a bad design.
    And this meant removing m_PcbFrame from BOARD.
  * removed BOARD::SetWindowFrame(), and BOARD::m_PcbFrame
  * Removed the global BOARD_DESIGN_SETTINGS which was in class_board.cpp
  * added BOARD_DESIGN_SETTINGS to the BOARD class, a full instance
  * a couple dialogs now only change BOARD_DESIGN_SETTINGS when OK is pressed,
    such as dialog_mask_clearance, dialog_drc, etc.
  * Removed common/pcbcommon.cpp's int g_CurrentVersionPCB = 1 and replaced it
    with build_version.h's #define BOARD_FILE_VERSION, although there may be a
    better place for this constant.
  * Made the public functions in PARAM_CFG_ARRAY be type const.
    void SaveParam(..) const and void ReadParam(..) const
  * PARAM_CFG_BASE now has virtual destructor since we have various way of
    destroying the derived class and boost::ptr_vector must be told about this.
  * Pass const PARAM_CFG_ARRAY& instead of PARAM_CFG_ARRAY so that we can use
    an automatic PARAM_CFG_ARRAY which is on the stack.\
  * PCB_EDIT_FRAME::GetProjectFileParameters() may no longer cache the array,
    since it has to access the current BOARD and the BOARD can change.
    Remember BOARD_DESIGN_SETTINGS are now in the BOARD.
  * Made the m_BoundingBox member private, this was a brutally hard task,
    and indicative of the lack of commitment to accessors and object oriented
    design on the part of KiCad developers.  We must do better.
    Added BOARD::GetBoundingBox, SetBoundingBox(), ComputeBoundingBox().
  * Added PCB_BASE_FRAME::GetBoardBoundingBox() which calls BOARD::ComputeBoundingBox()

Some code restyling: VECTOR_PCB is an array. Cleanup uncontrolled definitions of abs, max, etc. max is now overloaded function and MAX is a macro.

Metric KiCad work continues. Partially processed D_PAD class. It is still need to be tested including all these import and export procedures...

* All header files used to create the PCB common library now compile as
  stand alone code.  This prevents the need to define them in a specific
  order to make source code compile properly.  It should also now be
  possible to relocate the source code to build the common PCB library
  to a separate folder.

* Remove sine and cosine look up tables from trigo.cpp and replace them with
  sin() and cos() math functions.
* Large include file clean up to prevent header ordering dependency build
  failures.
* Translate French code names and comments.
* Lots of coding style policy and doxygen comment fixes.

* Changed <wx-2.8/xml/xml.h> to "xnode.h" in pcbnew_config.cpp to fix bug
  when building against wxWidgets 2.9 and above.
* Convert broken wxXmlNode code to use XNODE.
* Overloaded XNODE constructor for creating child nodes.
* Translate French naming conventions.
* Translate French comments.
* Remove tabs from several source files.
* Coding style policy and Doxygen comment fixes.

* Remove most (not all) variables that were initialized but never used that
  cause GCC 4.6 to generate a warning.
* Fix bug in schematic library editor best zoom calculation.

All: fix a minor bur annoying issue: incorrect cursor shape after a block command (mainly when a tool is selected)

* Translate source files D through L in pcbnew.
* Add translate tag for comments that did not translate correctly.

Added: Zones unfill in polygon mode