catch exception if data set is too large for SLIM

[zenogantner]

Thanks to Marcel for reporting this.

`
training data: 96556 users, 30924 items, 5852445 events, sparsity 99.804
test data: 5000 users, 27564 items, 155926 events, sparsity 99.88686
LeastSquareSLIM reg_l1=0.01 reg_l2=0.001 num_iter=15 K=50

---

* An uncaught exception occured. Please send a bug report to
zeno.gantner+mymedialite@gmail.com,*
*** or report the problem in our issue tracker:
https://github.com/zenogantner/MyMediaLite/issues ***
*** MyMediaLite 3.10
***

---

Array dimensions exceeded supported range. at
MyMediaLite.ItemRecommendation.SLIM.InitModel()
at MyMediaLite.ItemRecommendation.LeastSquareSLIM.InitModel()
at MyMediaLite.ItemRecommendation.LeastSquareSLIM.Train()
at ItemRecommendation.Train()
at MyMediaLite.Wrap.MeasureTime(Action t)
at ItemRecommendation.Run(String[] args)
at ItemRecommendation.Main(String[] args)
Terminate on unhandled exception.

training data: 96556 users, 30924 items, 5852445 events, sparsity 99.804
test data: 5000 users, 27564 items, 155926 events, sparsity 99.88686
BPRSLIM reg_i=0.0025 reg_j=0.00025 num_iter=15 learn_rate=0.05
uniform_user_sampling=True with_replacement=False update_j=True

---

* An uncaught exception occured. Please send a bug report to
zeno.gantner+mymedialite@gmail.com,*
*** or report the problem in our issue tracker:
https://github.com/zenogantner/MyMediaLite/issues ***
*** MyMediaLite 3.10
***

---

Array dimensions exceeded supported range. at
MyMediaLite.ItemRecommendation.SLIM.InitModel()
at MyMediaLite.ItemRecommendation.BPRSLIM.Train()
at ItemRecommendation.Train()
at MyMediaLite.Wrap.MeasureTime(Action t)
at ItemRecommendation.Run(String[] args)
at ItemRecommendation.Main(String[] args)
Terminate on unhandled exception.
`