Tricore EA calculation

[Rot127]

Your checklist for this pull request

• [ ] I've documented or updated the documentation of every API function and struct this PR changes.
• [ ] I've added tests that prove my fix is effective or that my feature works (if possible)

Detailed description

Fixes two issues with address calculation:

Fix absolute address calculations.

Absolute effective addresses are decoded by DecodeBInstruction(). Because they are encoded as normal disp24 value. To form the EA: The lower 20bits are shifted by one, and the upper 4bits by 1 (jumps) or 7 (calls).

Fix PC relative disponents.

The ISA sign extends disponents values of 8, 15 and 24 bits. For address disponents it also shifts the results by 1 for alignment.

The ISA has two writing styles for this though: sign_ext(dispXX 2) and sign_ext(dispXX) 2.

Assuming that sign_ext() always works on the MSB of the value they are eqivalent. Because: sign_ext(disp8 2) = sign_ext(disp8 2 = disp9) = sign_ext(disp9) = sign_ext(disp8) * 2.

This let to confusion before and for the sign_ext(dispXX * 2) case, and the wrong bit was checked for sign extension (e.g bit 8 for an effective disp9 value).

Test plan

Not yet added.

Closing issues

...

[Rot127]

@imbillow Please review this one before I fix the tests.

[XVilka]

@kabeor please merge this