Simple things to improve support for timestamped data

I often work with timestamped IOT data where I want to plot the number of events per day (or per hour or per week etc). There are a number of frustrations with this....

* The default behavior when using a DateTime field is to use the year/quarter/month/day hierarchy. I NEVER want to do this. Please let me turn off this behaviour.

* Most charts don't have any way to bucket/aggregate data directly. Surely it wouldn't be too hard to introduce an "aggregration period" on a time-based axis? Since our timestamps are to the nearest millisecond, the lack of aggregation effectively means that every single row is treated as a new data point to be plotted.

* I can work around the lack of aggregation by creating a new calculated column but DAX processing of DateTimes is very poor. In order to round a date to the nearest day I have to do a whole load of unpacking of individual fields or transform to seconds then back to a date. Really I just want to be able to write NewDate=DATEROUND(..column..,"12 hours") or similar.

* Related to above, it's absolutely ridiculous I have to repeat this for every new report I generate. Please provide a way to store a common library of user-defined DAX functions/macros that can be reused across reports.