Added dtemp and "m" to hfet1, pole-zero analysis and corrected a bug in unsetup function. Added two test netlists.

tests/hfet/id_vgs.cir

@@ -0,0 +1,13 @@
+HFET Id versus Vgs characteristic
+
+z1 1 2 0 hfet l=1u w=10u
+vgs 2 0 dc 0.3
+vds 1 0 dc 1.0
+
+
+.model hfet nhfet level=5 rdi=0 rsi=0 m=2.57 lambda=0.17
++ vs=1.5e5 mu=0.385 vt0=0.13 eta=1.32 sigma0=0.04
++ vsigma=0.1 Vsigmat=0.3 js1s=0 js1d=0 nmax=6e15
+
+.dc vds 0 1 0.01
+.end

tests/hfet/inverter.cir

@@ -0,0 +1,25 @@
+DCFL inverter circuit
+
+.subckt  inv 1 2 3
+*
+*Vdd 1.0
+*Vin 2 0
+*Vout 3  0
+z1 1 3 3   aload l=1u w=10u
+z2 3 2 0   adrv l=1u w=10u
+.ends
+
+vdd  1 0 dc 2
+vin  2 0 dc 0 pwl(0,0V   1ns,0V 1.005ns,1V 2ns,1V)
+x1  1 2 3 inv
+x2  1 3 4 inv
+.model adrv nhfet level=5 rd=60 rs=60 m=2.57 lambda=0.17
++ vs=1.5e5 mu=0.385 vt0=0.3 eta=1.32 sigma0=0.04
++ vsigma=0.1 vsigmat=0.3 js1s=1e-12 js1d=1e-12
++ nmax=6e15
+.model aload nhfet level=5 rd=60 rs=60 m=2.57 lambda=0.17
++ vs=1.5e5 mu=0.385 vt0=-0.3 eta=1.32 sigma0=0.04
++ vsigma=0.1 vsigmat=0.3 js1s=1e-12 js1d=1e-12
++ nmax=6e15
+
+.end