Ask 1: How can I avoid the spike signal by choosing the segments below the inclusion signal?

Bence: If you go to the Tools menu -> Automatic Selections -> From Channel Data, you'll get the window in the screenshot below. There you just need to enter criteria that match your sample, but not your inclusions. For example, if your inclusions have high Sr, but low Si, you can add criteria for low Sr and higher Si. The actual values you'll need to determine from your data, but that's the way to approach it: select what is normal for your samples.

It's also very useful to calculate the ppm channels first and use those as they are not affected by signal drift or baselines.

Ask 2: So I guess this approach is also suitable for a suite of analysis, not limited to a single one. Right?

Bence: Yes, exactly! You can set up 'recipes' to avoid all sorts of inclusions (e.g. apatite inclusions in zircons). Or you can go the other way, and create a recipe for each mineral. The thing to be aware of though is that you don't want to use any elements that vary naturally in your minerals, otherwise you might start to reject natural variation in your samples.

Ask 3: If I want to check the channels of each analysis for samples after executing DRS to see whether the dwell time is suitable for each analysis, which channels should I refer to, the inputs, the intermediates or the outputs?

Bence: Yes, I would look at the uncertainty on your output channels for your secondary reference materials. Choosing dwell times can be tricky: it is a trade-off between good reproducibility for each channel, but also you want good time resolution so that you don't get very smoothed signals. I would look at it from the perspective of "What precision do I need for each element?" And then start adjusting the dwell times until you get that precision, with the smallest total sweep time possible.