[r13472]

 - number of interfering affinity adres was calculated wrong
 - recoloring stopped after first chunknode, must be run till end

Output more like heur3

Seems to work now (al leat Queens produces identical result).

[r13471]

[r13470]

[r13464]

[r13463]

removed some unnecessary code

[r13462]

removed #if 0'ed functions

[r13461]

[r13454]

[r13446]

Fixed: Removed wrong warnings regarding strict Conv's
Strict Conv's still not implemented!

[r13445]

decorate_win32_c_fkt() now gets the initial id as a parameter instead reading it from the entity itself.
This simplifies some Java-frontend code.

[r13444]

Currently non-symmetric!

[r13443]

[r13442]

[r13441]

[r13434]

[r13433]

fixed several bugs

[r13432]

assert that in and out entities of memperms have the same size, we produce such buggy situations at the moment

[r13431]

[r13430]

[r13426]

[r13425]

[r13423]

[r13422]

[r13421]

[r13420]

[r13419]

[r13418]

[r13417]

some indents changed and comments added

[r13416]

[r13415]

[r13414]

[r13413]

[r13412]

added missing void declaration

[r13411]

added comment

[r13410]

[r13404]

[r13401]

[r13400]

[r13399]

[r13398]