WIP: Two stage method

[shivin9]

An attempt to implement the two-stage method without using Optim.jl. Still it has some dimentionality issues though.
I am using curve_fit to find the optimal parameters beta_0 and beta_1.
@ChrisRackauckas do you think this can work?

[coveralls]

Coverage decreased (-27.4%) to 54.762% when pulling 8b7a946 on shivin9:two_stage_mthd into bb0c788 on JuliaDiffEq:master.

[codecov-io]

Codecov Report

Merging #7 into master will decrease coverage by 27.38%.
The diff coverage is 0%.

@@             Coverage Diff             @@
##           master       #7       +/-   ##
===========================================
- Coverage   82.14%   54.76%   -27.39%     
===========================================
  Files           1        1               
  Lines          56       84       +28     
===========================================
  Hits           46       46               
- Misses         10       38       +28

Impacted Files	Coverage Δ
src/DiffEqParamEstim.jl	54.76% <0%> (-27.39%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update bb0c788...8b7a946. Read the comment docs.

[ChrisRackauckas]

See #4 (comment)

[ChrisRackauckas]

Closing this because an implementation was just merged. However, there's still some to be done along these lines, like #10 .