update readme

README.md

@@ -5,3 +5,15 @@
 *A fast, multi-system, multi-kernel, differentiable implementation of the fast multipole method for use with scalar-plus-vector potential N-body problems in pure Julia.*
 
 Authors: Ryan Anderson and contributors
+
+Features:
+
+* Laplace $1/r$ kernel
+* meant to incorporate seamlessly into existing Julia code without modifications (only additions to define a few interface functions)
+* convenience functions for determining the expansion coefficients for source points, vortex points, source panels, and dipole panels (this list is growing!)
+* $\mathcal{O}(p^4)$ multipole-to-local translation operator (where ```math p``` is the expansion order), though this may improve in the near future
+* velocity and velocity gradient (or their equivalent for non-fluids problems) obtained using analytic expressions (no finite difference)
+* CPU-parallelization for expansions and direct interactions
+* GPU-parallelization for direct interactions using `CUDA.jl`
+* `ForwardDiff` and `ReverseDiff` compatible
+