It sounds easy but there are some inherent consequences that may not be that common to be found in modules but they do exist (I have made use of that trick more than once): Sometimes there is silence at the end of the sample just so that it keeps playing and making it possible to retrigger it (in IT the sample will not retrigger if it has stopped playing). In this case the Automatic Sample Trimmer will then do the only right thing that it can do in that situation: It recognizes that the silence is being played, and doesn't trim it. In MOD there could be similar consequences due to the on-the-fly sample swapping mechanics. Hence I'm rather inclined not to have this optimization. Too many things can go wrong without the user realizing it before it's too late.

Also, trimming silence at the start of the sample will obviously change the sound of the module because now the sampled sound will be heard earlier than intended. This is also found in real-world modules e.g. in pre-shifted claps (clap samples that have a fade-in before the actual clap is being heard, with silence at the beginning to make them align properly to the beat grid). This is something that absolutely shouldn't be decided automatically.

Oh, I didn't realize that. I realized that trimming silence at the start shouldn't be done automatically, which is why I said it could be included in the Cleanup as a popup. Perhaps make trimming silence at the end as a popup asking you whether you want to or not?

But I do agree now that there are many things that can go wrong with this.

The cleanup dialog is simply the wrong place for this. Its intention is to never modify the module in an audible way - it just cleans up the module, it doesn't optimize the last bits out of it. The peresence of 100% silence is also only really relevant for 8-bit samples most of the time - because quite often, you will find that what looks like total silence is in fact still a bit of noise in 16-bit samples. I'd say that in either case it's still best if the musicians goes through the samples one by one to make their decisions. However, that doesn't mean that there's nothing that OpenMPT can help with - for example there could be a sample editor tool that detects and selects silence - then you can do with that selection whatever you want.