compiler: rewrite jump targets properly

[fyrchik]

Closes #630 .

Old implementation could view 0x62 byte in
a script as a JMP instruction irregardless of whether it is
a real opcode or a part of a parameter of another instruction.
In this commit instructions are decoded together with parameters
during jump label rewriting.

[codecov]

Codecov Report

Merging #643 into master will decrease coverage by 0.02%.
The diff coverage is 77.77%.

@@            Coverage Diff             @@
##           master     #643      +/-   ##
==========================================
- Coverage   65.07%   65.04%   -0.03%     
==========================================
  Files         125      125              
  Lines       10569    10572       +3     
==========================================
- Hits         6878     6877       -1     
- Misses       3413     3416       +3     
- Partials      278      279       +1

Impacted Files	Coverage Δ
pkg/vm/context.go	70.93% <0%> (-1.69%)	⬇️
pkg/compiler/codegen.go	86.6% <100%> (-0.28%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update cc9ace6...3e84f2b. Read the comment docs.

[roman-khimov]

I'm not sure why do you need another Decode function? Just (re)using NewContext() and ctx.Next() should be enough, no?

[fyrchik]

I'm not sure why do you need another Decode function? Just (re)using NewContext() and ctx.Next() should be enough, no?

I need to know how much bytes were actually read. Decode has somewhat lower level interface. And it seems cleaner to have it and reuse.

[roman-khimov]

Hmm... I'm not convinced, I think it's easier to add some NextIP() method to Context and be done with it.