I think there are some more general underlying questions in this, implied and necessarily precedent to the question asked. Answering them, answers the question.
- Is a supervisor bound to entertain every question a researcher might ask?
I'd argue no.
I feel it's ethically fine for a supervisor to set well-defined boundaries. An extreme might be "no more than 10 questions a day, by email, before 10am. In each, show what reasonable effort you made to find the answer yourself" - I've laid down similar boundaries myself, albeit far less stringent, when mentoring large numbers of people.
It'd also be ethically fine for a supervisor to fail to respond to questions which were clearly asked to large numbers of people at once; were abusive or personal; were asked through a weird medium; were not on the topic of the research; or were otherwise in violation of commonly-implied social boundaries for expecting a question response.
So the supervisor is not ethically required to even respond to every question from a researcher.
- If an appropriate question is appropriately asked, is the supervisor bound to respond?
I'd argue yes.
If the researcher has shown reasonable effort in their search for answers, and found none, then they've likely encountered a "blocker".
It's arguably the whole point of mentoring, to hear and respond to problems encountered by those we're helping: without such interactive guidance, we're just a YouTube video on legs.
Are there those who won't fulfill this ethical responsibility? Absolutely. There are crap supervisors in academia, just as there are crap managers in business. And unless there's some explicit rule in place in the institution you're working at to force them to take a greater interest in helping, there may not be a lot you can do about that.
- What form of response is the supervisor bound to give?
Any of the following would be fine:
- "I don't know how to help you with that."
- A partial answer, acknowledging that it's only partial.
- A full and complete answer of the question, maybe even raising additional issues that might be relevant.
- Proposing a reframe of the question, if it seems based on a deeper knowledge failure. Arguably the best help a mentor can give is to see beyond simple questions to deeper issues.
- A pointer to likely sources of the answer.
- A suggestion of other avenues of inquiry or research approaches that might lead to an answer.
- ... and so would many other approaches.
And as others are said, answering every potential question is neither the only, nor the best response to every question anyway, since we're trying to teach people to fish, not give them a coupon for unlimited free fishsticks.
I suspect that's why the OP didn't say "answer", but rather "clarify".
TL;DR: For any reasonable question, asked reasonably, and within any explicit boundaries set by the supervisor, YES, but only a response, not necessarily an answer. It's a mentor's role to give support and advice.