I'm sure there are people who are doing it, but I see them as being more relevant to OLTP at the moment than OLAP.
In my opinion, in a data warehouse you're mainly looking for:
High sequential reads
Lots of storage capacity
Lots of memory
Although SSD's have high throughput rates, it's not the same order of magnitude higher than HDD's as it is for IOPS/random reads/writes. Therefore, you're likely to find a better price/performance rate from a large array of 15k SAS in Raid 10 than you could for SSD's, especially if you're looking at multi-terabyte DB's.
As the profile is generally high on reads rather than writes, you're also better off scaling up memory as you don't have the same I/O barrier as you do for OLTP workloads.