fix: validate that token name doesn't contain only spaces

[bassgeta]

Description

Added a validation test to check if the string isn't just whitespaces. I however allow for leading and trailing whitespaces, but they get trimmed out when creating the token? 🤷

Testing

1. Start up your dev env and run node ./scripts/create-colony-url.js
2. Select "Create a new token"
   
3. Now try to break the token name validation, entering only spaces should fail
   
   But spaces at the start/end should work if you have any content
   

Diffs

Changes 🏗

• createTokenValidationSchema now includes whitespace only validation for tokenName

Resolves #3298

[iamsamgibbs]

Nice job this is working as expected!

I guess we need a product decision on whether leading / trailing spaces?

Either way it looks like we'll need to address how the colony name is displayed with spaces in elsewhere in the app as it doesn't deal well with multiple spaces:

It does retain the spaces in the db though, so this is purely a UI issue:

None of that prevents this from getting merged though!

[rdig]

All good here, it now validates against empty spaces

However I can still bypass it with Unicode spaces (in this case U+0020)

This should not make the focus of this PR, and arguably we shouldn't bother with Unicode validates as most nobody does (just test this on any major service's forms, and you won't get validated against -- this is actually my go to when I don't want to provide a name 😛).

However one could also argue that since this is a Token and it's going to be deployed "forever" with that name, it might become confusing, so we might want to prevent against that. (maybe run this validation through our "confusable characters" lib that we use for usenames?)

PS: for testing unicodes I use this website https://www.editpad.org/tool/invisible-character