Implement missing RegExp properties

[RageKnify]

ECMASCript feature
The description of the RegExp Constructor is here: https://tc39.es/ecma262/#sec-regexp-constructor

Example code
MDN should have enough examples to test, https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp

• get RegExp [ @@species ]
• RegExp.prototype.constructor
• RegExp.prototype [ @@match ] ( string )
• RegExp.prototype [ @@matchAll ] ( string )
• RegExp.prototype [ @@replace ] ( string, replaceValue )
• RegExp.prototype [ @@search ] ( string )
• get RegExp.prototype.source
• RegExp.prototype [ @@split ] ( string, limit )

Part of the code for @@replace is in String.prototype.replace and should be removed from there.

boa/boa/src/builtins/string/mod.rs

Line 648 in 2e75dd8

pub(crate) fn replace(this: &Value, args: &[Value], context: &mut Context) -> Result<Value> {

[raskad]

I am currently working on this.

[raskad]

Depending on how we want to define this issue, I think we could mark some items on the list as done or close this issue.
The following is the current state of the built-ins/RegExp/prototype test suite:

Test Suite	Total	Passed
RegExp.prototype	880	713
Symbol.species	8	8
Symbol.match	102	100
Symbol.matchAll	52	34
Symbol.replace	134	105
Symbol.search	46	38
source	24	12
Symbol.split	88	54

My feeling is that a big part of the failing tests is based on missing features, either somewhere in boa or in regress. For the most part, I think the basic features are implemented.

[RageKnify]

Depending on how we want to define this issue, I think we could mark some items on the list as done or close this issue.

My opinion is that if a function is spec compliant we can consider it done.

Tests that depend on regress implementing more things could be open as separate issues and include a link to regress's issue so that we have something tracking it here.

If the failing tests depend on things being implemented elsewhere in boa (for example Reflect is used in a lot of tests outside its folder) then we can also consider that function done, and it might be a good idea to open an issue for the missing feature if none exist.

[raskad]

I've looked trough all of the failing tests for the RegExp/prototype tests. Here is a list of the related feature / bugfix that is necessary for the tests to pass.

Feature / Bug	Failing test
ComputedPropertyName in ObjectLiteral get	Symbol.matchAll/isregexp-this-throws Symbol.matchAll/isregexp-called-once Symbol.matchAll/species-constructor-get-species-throws
new Object("a").toString() does not return 'a'	exec/S15.10.6.2_A1_T3
RegExp Match Indices #1428	flags/return-order flags/rethrow flags/coercion-hasIndices flags/get-order flags/this-val-regexp
Implementing the eval function #948	source/value source/value-line-terminator source/value-slash source/value-empty source/value-u
Realm related?	Symbol.split/splitter-proto-from-ctor-realm ignoreCase/cross-realm dotAll/cross-realm sticky/cross-realm multiline/cross-realm source/cross-realm unicode/cross-realm hasIndices/cross-realm global/cross-realm
Related to assigning this in function?	Symbol.replace/fn-invoke-this-no-strict
UTF16 surrogate not supported #736	Symbol.match/builtin-infer-unicode Symbol.replace/coerce-unicode
var initialized hoist with undefined currently not implemented	exec/S15.10.6.2_A4_T5 exec/S15.10.6.2_A1_T20 test/S15.10.6.3_A1_T20

[jedel1043]

Closing because #1434 solved most, if not all, the differences from the spec and we have all properties implemented for RegExp