I dont think this is a priority, but if this comes up as a high priority customer requirement, if lots of customers keep asking for this (voting for it in the ideas portal) then this solution i have seen elsewhere would be worth knowing about.
Basically, it's really hard to do a group report on unstructured numeric values, especially those with decimals or a wide range, or dates, or even "random strings" that have no structure like customer names.
A good example might be trying to group quotes by value, where every quote has a unique value, or by age where you would get individual days.
You can solve the date one by using date "bucket" functions that are pretty common such as month(), year() etc. but probably need to be careful of the performance impact of computing those on the fly.
For value however there isnt anything quite like this. I have seen in a few systems the ability to group by "value Bucket" - which may be a bit dynamic (in terms of how big the buckets are) but seems to be like a lookuplist of pre-defined ranges. So, Quote might have a field of "ValueBucket" that links to a lookup list of "ValueBucketRanges" - i.e. 0-1000, 1000-2000, 2000-5000, 5000-10k, 10k+
See attached image to see an example of this applying to "duration" - which is a decimal and so really doesnt group well.