It's actually kind of a hard problem.

There are tens of thousands of grants. Some of them are classified as "medicine" and study things like hormones, some are classified as "computer science" or "mathematics" and study things like statistical bias. Which of these are the "real science" ones (e.g. HRT in older people with menopause or low-T) and which are the partisan-coded ones (e.g. gender HRT)?

Put aside whether or not you think the latter should be funded and suppose you're just trying to distinguish them to make sure the former continues to be. If you can't do it accurately then the current administration will do it anyway but with a machete instead of a scalpel. You'd need a large team of people who can tell the difference to go through them all and classify them. This is nominally the job of the thing referred to as the "deep state", but what do you do if you don't expect them to be non-partisan?

This is why strategies like "have partisans capture the administrators for our side" are a mistake.

I'm curious why you think this would be an easy task. It strikes me as quite difficult! Separating scientific wheat from chaff requires a lot of effort and expertise.

As one famous example, Sokal intentionally made up funny bullshit and still fooled a ton of people!

Research on weather and climate is now considered "woke". Anything can be labeled as partisan. Safer to stop funding all research, unless it can be shown to support what we already believe.