[feature/refactor] general bugs #165

I'm going to use this issue to track all the bugs I find while working on the LSP

Error Reporting

• missing call argument reported incorrectly
  This should be reported as too many call arguments or better 1 parameter expecetd but 2 were given

block.setFriction(1, "");

This happens here

• call argument error position is incorrect
  The position should be reported in between the ()
  
• Unknown variable has a weird reporting message, this can probably be more vague, like cannot find X
• (opinion) ambigious call is too verbose

var b = a.map<string>(x => "a" + x);

is reporting as

Ambiguous call; multiple headers match:
<U@Arrays.zs:50:12>(projection: function(value: string): string): string[]
<U@arrays_map_2.zc:7:12>(projection: function(value: string): string): string[]

Could be reported as:

Ambiguous call; multiple headers match:
<U@Arrays.zs:50:12>(function(string): string): string[]
<U@arrays_map_2.zc:7:12>(function(string): string): string[]

There could be more cleanup, but I think removing the parameter names are a good start
Java reports this error as:

Ambiguous method call: both 'CompilerErrors.doThing(Function<String, String>)' and 'CompilerErrors.doThing(int[])' match.

• No operator in type position is incorrect when in call arguments
  The position should be reported on the operator, instead of on the function call
  

• Iterator errors are reported in the wrong place
  

General

• If a class has a constructor, the ContructorMember's body never gets set and is always null
  

  ZenCode/CodeModel/src/main/java/org/openzen/zenscript/codemodel/definition/ClassDefinition.java

  Line 46 in f325e2f

  boolean hasNoConstructor = members.stream().noneMatch(IDefinitionMember::isConstructor);