Avoid race between updating stepper and ViewPager instantiating the steps

[bthrall]

It was assuming the view pager has layed out all the steps at the time that it calls back to its listeners,
but that is not always true (Handler.post() is happening before ViewPager has inserted the step fragments into the
fragment manager).

This solution waits for the ViewPager to complete its layout before updating StepperLayout.

[codecov-io]

Codecov Report

Merging #127 into master will decrease coverage by 11.26%.
The diff coverage is 50%.

@@             Coverage Diff              @@
##             master    #127       +/-   ##
============================================
- Coverage     46.37%   35.1%   -11.27%     
+ Complexity      113      86       -27     
============================================
  Files            24      24               
  Lines           882     883        +1     
  Branches         85      85               
============================================
- Hits            409     310       -99     
- Misses          409     531      +122     
+ Partials         64      42       -22

Impacted Files	Coverage Δ	Complexity Δ
...main/java/com/stepstone/stepper/StepperLayout.java	24.34% <50%> (-9.66%)	11 <0> (-6)
...stepstone/stepper/internal/util/AnimationUtil.java	0% <0%> (-57.9%)	0% <0%> (-2%)
.../com/stepstone/stepper/internal/util/TintUtil.java	8.82% <0%> (-47.06%)	2% <0%> (-3%)
...e/stepper/adapter/AbstractFragmentStepAdapter.java	55.55% <0%> (-22.23%)	2% <0%> (-1%)
...epstone/stepper/internal/widget/TabsContainer.java	68.25% <0%> (-14.29%)	13% <0%> (-4%)
...com/stepstone/stepper/internal/widget/StepTab.java	29.57% <0%> (-13.39%)	12% <0%> (-3%)
...epstone/stepper/internal/type/DotsStepperType.java	60% <0%> (-13.34%)	2% <0%> (-1%)
.../stepper/internal/type/ProgressBarStepperType.java	56.25% <0%> (-12.5%)	2% <0%> (-1%)
...com/stepstone/stepper/viewmodel/StepViewModel.java	47.05% <0%> (-11.77%)	2% <0%> (-4%)
...epstone/stepper/internal/type/TabsStepperType.java	65.38% <0%> (-11.54%)	4% <0%> (-1%)
... and 2 more

---

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 5cffbfe...abb2c4d. Read the comment docs.

[zawadz88]

Hi,
Thank you for this PR, we'll review it shortly :)
Could you provide steps to reproduce the existing issue so that we can try to write tests against it, please?

[zawadz88]

Ok, so I was able to reproduce this and write some tests against this (I'll add them once this is merged).
Thanks for fixing this!