List-2018-Colexifications & Rzymski-2020-Colexifications

[MiraAhmedovic]

Adding Network adaptations of List-2018-Colexifications and Rzymski-2020-Colexifications.

In Rzymski, ENGLISH and CONCEPTICON_GLOSS were refined so that they are identical except for the cases where they shouldn't be:

CONCEPTICON_GLOSS	ENGLISH
YOUNG GOAT (KID)	KID
PATERNAL UNCLE (FATHER'S BROTHER)	FATHER'S BROTHER
RAISE (BRING UP)	RAISE (A CHILD)
FAN (ACTION)	FAN ACTION
BOAR (MALE PIG)	BOAR
MORTAR	MORTAR CRUSHER
KNOW HOW	KNOW OR BE ABLE

[MiraAhmedovic]

The IDs in Rzymski only have 3 parts (Name-Year-Number instead of Name-Year-NoItems-Number), which the metadata aren't too happy with, since they expect 4 elements:
"datatype": {
"base": "string",
"format": "[a-zA-Z]+\-[0-9]{4}\-[0-9]+[a-z]?\-[0-9]+[a-z]?$"
},
"name": "ID"

If the IDs in Rzymski are supposed to stay only 3 elements long, the metadata will have to be adjusted:
{
"datatype": {
"base": "string",
"format": "[a-zA-Z]+\-[0-9]{4}\-[0-9]+[a-z]?$"
},
"name": "ID"

If that's the case then everything runs smoothly.

[LinguList]

@AnnikaTjuka, if you also agree, please just merge this :-)

[AnnikaTjuka]

Sorry, this slipped my mind. Looks good, I'll merge it.