feat(databricks)!: support comma-separated syntax for OVERLAY function

[AbhishekASLK]

Previously, the parser defaulted to the strict ANSI implementation, causing a ParseError when encountering the
comma-separated variant.

SQL:

SELECT overlay('Spark SQL', 'ANSI ', 7, 0)

Databricks Docs

[VaggelisD]

Thank you for the contribution @AbhishekASLK! Two minor comments to consider and its fine to merge

[VaggelisD]

We can probably simplify this as:

        def _parse_overlay(self) -> exp.Overlay:
            def _parse_overlay_arg(text: str) -> t.Optional[exp.Expression]:
               return (self._match(TokenType.COMMA) or self._match_text_seq(text)) and self._parse_bitwise()
             
            return self.expression(
                exp.Overlay,
                this=self._parse_bitwise(),
                expression=_parse_overlay_arg("PLACING"),
                from_=_parse_overlay_arg("FROM"),
                for_=_parse_overlay_arg("FOR"),
            )

[abhishek-bhosale-ksolves]

Thanks to simplify this, will modify it as suggested.

[VaggelisD]

validate_all should be used when we also want to check transpilations between other dialects; When testing the same dialect with expected changes we should still use validate_identity:

Suggested change

	self.validate_all(
	"SELECT OVERLAY('Spark SQL', 'ANSI ', 7, 0)",
	write={
	"databricks": "SELECT OVERLAY('Spark SQL' PLACING 'ANSI ' FROM 7 FOR 0)",
	},
	)
	self.validate_identity(
	"SELECT OVERLAY('Spark SQL', 'ANSI ', 7, 0)",
	"SELECT OVERLAY('Spark SQL' PLACING 'ANSI ' FROM 7 FOR 0)"
	)

[abhishek-bhosale-ksolves]

Okay, got it!