`sysret` should be considered a terminator

[whitequark]

Version and Platform (required):

• Binary Ninja Version: 5.0.7208-dev Personal (9967e4cb)
• OS: Windows
• OS Version: 10
• CPU Architecture: x64

Bug Description: sysret sets RIP to RCX and so it should be considered a function terminator (not merely a basic block terminator, as it currently is, with a fallthrough to the next instruction).

Steps To Reproduce:

1. Open sysret.bin
2. Create an x86_64 function at offset 0

Expected Behavior: A function with two instructions: swapgs; sysret.

Screenshots/Video Recording: This is what it currently looks like:

Binary: sysret.zip

[plafosse]

Ok this actually looks like its implemented as an intrinsic that sets RIP this is clearly wrong.

[whitequark]

Thank you!