VPU: new vcompress to fit v0&vl split; fix vfredsum/min/max

[lewislzh]

fix vfredsum/max/min: When the vector vfredusum/max/min consists entirely of inactive elements and vs1[0] is NaN, the result should be vs1[0] When both elements of vfredusum are inactive, the temporary result changes from positive zero to negative zero. nes vcompress to fit v0/vl split: The vcompress calculation combines the ones_sum result with vs1 using a temporary register to reduce one read operation. Additionally, other uops, except ones_sum, reduce the basemask calculation and the right shift basemask operation. fix vpermtest to fit new vcompress

[XiangShanRobot]

[Generated by IPC robot] commit: da57bb3fd6bd2c42dca067d3e82ff6ef0392a620	commit	astar	copy_and_run	coremark	gcc	gromacs	lbm	linux	mcf	microbench	milc	namd	povray	wrf	xalancbmk
da57bb3	1.827	0.448	2.048	1.190	2.938	2.508	2.289	0.932	1.369	1.410	3.445	2.672	2.398	2.932

master branch:	commit	astar	copy_and_run	coremark	gcc	gromacs	lbm	linux	mcf	microbench	milc	namd	povray	wrf	xalancbmk
ef14270	1.827	0.448	2.048	1.190	2.938	2.508	2.289	0.932	1.369	1.410	3.445	2.672	2.398	2.932
5c5f442	1.827	0.448	2.048	1.190	2.938	2.508	2.289	0.932	1.369	1.410	3.445	2.672	2.398	2.932
d67c873	1.827	0.448	2.048	1.190	2.938	2.508	2.289	0.932	1.369	1.410	3.445	2.672	2.398	2.932
00f9d18	1.809	0.448	2.060	1.191	2.938	2.508	2.290	0.932	1.419	1.338	3.431	2.642	2.398	2.932
3b94d5d	1.809	0.448	2.060	1.191	2.938	2.508	2.290	0.932	1.419	1.338	3.431	2.660	2.398	2.932
2f6c010	1.820	0.448	2.060	1.199	2.938	2.508	2.290	0.932	1.419	1.338	3.427	2.651	2.398	2.932
0f42355	1.820	0.448	2.060	1.199	2.938	2.508	2.290	0.932	1.419	1.338	3.427	2.651	2.398	2.932
95e6033	1.820	0.448	2.060	1.199	2.938	2.508	2.290	0.932	1.419	1.338	3.427	2.651	2.398	2.932
58cb1b0	1.822	0.448	2.060	1.180	2.944	2.503	2.291	0.932	1.419	1.328	3.437	2.644	2.399	2.931
202ef6b	1.815	0.448	2.060	1.182	2.953	2.504	2.291	0.930	1.403	1.319	3.426	2.660	2.397	2.940