Unicode chars are ignored as enum values

Generating java code from spec with enums using non-latin (russian-cyrillic) chars gives invalid output. For example for enum like:

"enum" : ["один", "два", "три"] ...

output (for java) looks like:

public enum ... {
  _("один"),
  _("два"),
  _("три")
 ...
}

Which gives compile error. Since java supports unicode chars for enum names - corresponding code should be replaced with one which will not trim non-latin chars.

https://github.com/OpenAPITools/openapi-generator/blob/8f5639554e96a9beaf0b4c87b9dac52afcc1ef83/modules/openapi-generator/src/main/java/org/openapitools/codegen/languages/AbstractJavaCodegen.java#L1406

Proposal: replace regexp with Unicode-enabled one

String var = value.replaceAll("\\W+", "_").toUpperCase(Locale.ROOT);

with

String var = value.replaceAll("(?U)\\W+", "_").toUpperCase(Locale.ROOT);

Start in modules/openapi-generator/src/main/java/org/openapitools/codegen/languages/AbstractJavaCodegen.java at the enum-name handling around line 1406. Generate Java code from a spec containing Cyrillic enum values and verify the generated enum constants preserve those characters and compile successfully.