fix: Implemented whole word matching in ProfanityAnalyser class

[gabrielrbarbosa]

Example in portuguese:
'Colar de pérola' would get incorrectly flagged as profanity because it contains the last 4 chars that is a bad word in pt_BR config.
After this fix: PASS No profanity found in your application!

\p{L} matches any Unicode letter.
(?<!\p{L}) ensures the character before isn’t a letter.
(?!\p{L}) ensures the character after isn’t a letter.
u flag = Unicode mode
i flag = case-insensitive

[gabrielrbarbosa]

Also one thing unrelated to this PR but I think could be removed is "sheet" from en config, because it is a common word for spreadsheet export right?

pest-plugin-profanity/src/Config/profanities/en.php

Line 1338 in 06c77bb

'sheet',

[JonPurvis]

Thanks for the contribution @gabrielrbarbosa

Are you able to take a look at the failing static analysis checks please? Happy to merge this once resolved.

[gabrielrbarbosa]

@JonPurvis fixed style issue

[JonPurvis]

Thanks for sorting! Interesting that you mentioned sheet, I had that crop up on a project yesterday! I ended up just adding --exclude=sheet into my composer script which did the trick. I'll have a think about whether we should just exclude it in the package core though.