Implement finite field ccopy, neg, cneg, nsqr, ... for CUDA target
#2127
| Job | Run time |
|---|---|
| 9s | |
| 6s | |
| 13s | |
| 7s | |
| 34m 50s | |
| 31m 56s | |
| 14s | |
| 8s | |
| 13s | |
| 8s | |
| 3m 57s | |
| 3m 41s | |
| 1h 15m 42s |