Column names

[joellepasman]

• In the output of ldsc_rg h2_trait_se should be h2_trait1_se to make the distinction with h2_trait2_se.
• For clarity, it could be considered to change p to rg_p to make clear that this is the p-value of the rg, not the h2.
• It would be nice to have an additional column containing the name of trait1.

[joellepasman]

Plus it would be nice if the code would automatically derive Z from B/SE if not given in the input