umxSummary: add p-value to estimate table

[tbates]

Modify umxSummary to include a p-value column based on beta and SE

libs("mlbench")
data(BostonHousing2)

BostonHousing2$log_crim  = log2(BostonHousing2$crim)
BostonHousing2$nox  = BostonHousing2$nox*100
m3 = umxRAM(data = BostonHousing2, "#crime_model
	cmedv ~ log_crim + b1*nox; 
	nox   ~ a1*rad + a2*log_crim
	i_1 := a1*b1
	i_2 := a2*b1"
)

Table: Parameter loadings for model 'crime_model'

	name	Estimate	SE	type
8	log_crim_with_rad	23.132	1.584	Manifest Cov
1	b1	-0.144	0.051	Manifest path
2	a1	-0.302	0.068	Manifest path
3	log_crim_to_cmedv	-0.925	0.188	Manifest path
4	a2	3.648	0.191	Manifest path
10	one_to_cmedv	29.470	2.995	Mean
11	one_to_nox	62.457	0.900	Mean
12	one_to_rad	9.549	0.387	Mean
13	one_to_log_crim	-1.126	0.139	Mean
5	cmedv_with_cmedv	65.499	4.118	Residual
6	nox_with_nox	48.798	3.068	Residual
7	rad_with_rad	75.667	4.757	Residual
9	log_crim_with_log_crim	9.710	0.610	Residual

Model Fit: χ²(1) = 0.08, p = 0.777; CFI = 1.001; TLI = 1.004; RMSEA = 0
Algebra'i_1' = 0.043CI95[0.008, 0.079]. p-value = 0.017
Algebra'i_2' = -0.525CI95[-0.89, -0.16]. p-value = 0.005

[mcneale]

While you're at it, how about +/-1.96 SE's for SE-based confidence intervals? This would be nice to conveniently compare with likelihood-based confidence intervals.