Should non-native programmers favor regular verbs over irregular to avoid incorrect naming?

There are two hard things in computer science, naming is one of them.

In this case, it's not so hard: Code should be easy to read, and the words used should not cause irritation, so using grammatically correct names is a good thing (although code will not follow the language grammar overall).

I would not go so far as to forbid irregular verbs for use by non-native speakers (how are they supposed to know whether a verb is irregular if their language proficiency is low?) but a healthy culture of improving code quality, including renaming and other actions which make the code more readable (preferably during review). If you use English words in code written by speakers who are prone to make mistakes at that level, the review should be done by someone who can spot and correct these mistakes.

1. leave it as it is
2. use correct English
3. try to avoid irregular verbs

I think all these are potential solutions!

Obviously the starting point should be English as we know it.

Then you potentially adjust the vocabulary to avoid tricky results, including grammatical infelicities.

And finally if the "wrong" spelling or grammar is at least part of a rational scheme, then consider permitting it, though I think it is more customary in programming to break grammar rules than the conventional spellings (for example, Bind() and bool IsBind, or Find() and bool FindDone, rather than bool Binded and bool Finded).

Over the years I've also come to disfavour plurals for arrays/collections precisely because of the inconsistencies they lead to (currency/currencies, money/monies, person/people, before you even get to questions like schemas vs schemata). It's usually better to suffix with "set" or "list".

Merely ad-hoc and unintentional spelling mistakes by a writer unskilful with English, shouldn't be accepted outside the context of local variables.

With level of details from description quoted here for reference...

In a few projects I've seen names like "foundedFile" and "bindedDatasource". What would be the industry consensus be: leave it as it is, use correct English "foundFile" / "boundDatasource", try to avoid irregular verbs in naming where they cause misunderstanding?

... it is opinion based. To get objective answers the context needs to be detailed since the naming could be intentional either since it models some real life use cases or it is just a form of obfuscation from some obscure security concerns that when are not of interest anymore the way to go is speak the language the development team understands and that shouldn't require approval or be stated by guide lines, is just grammar we all have been taught before being taught programming.

I think you have the same problem with native speakers and plurals

Person person
Person[] persons //should be people, personArray? for correct english
Sheep sheep
Sheep[] sheep //should be sheeps? for coding standard collection naming

English has a bunch of crazy rules and often there isn't a good choice for variable names.

I've also seen spelling mistakes persist throughout large code bases.

I think there is a good argument for incorrect english where it provides consistency in variable names. eg say you have a bunch of paired functions

func click
bool clicked

func swipe
bool swiped

func error
bool errored //erred is correct english

func find
bool finded //found is correct english

func flavor   //flavour is correct
bool flavored //flavoured is correct

Given this, am I going to correct peoples grammar when they are programming in a foreign language? Seems like a richard move.