Remove unnecessary rank check in potrs call.

[bamos]

Hi, this PR removes an unnecessary rank check in the potrs code.
To illustrate why I think this is unnecessary, consider the following example.

A = torch.Tensor({
    {1.2705,  0.9971,  0.4948,  0.1389,  0.2381},
    {0.9971,  0.9966,  0.6752,  0.0686,  0.1196},
    {0.4948,  0.6752,  1.1434,  0.0314,  0.0582},
    {0.1389,  0.0686,  0.0314,  0.0270,  0.0526},
    {0.2381,  0.1196,  0.0582,  0.0526,  0.3957}})

B = torch.Tensor({
    {0.6219,  0.3439,  0.0431},
    {0.5642,  0.1756,  0.0153},
    {0.2334,  0.8594,  0.4103},
    {0.7556,  0.1966,  0.9637},
    {0.1420,  0.7185,  0.7476}})

chol = torch.potrf(A)
torch.potrs(B, chol)
torch.potrs(torch.cat({B,B}, 2), chol)

cat(B,B) fails the rank check, but this is irrelevant and
shouldn't result in an error.

With the unnecessary check

th> torch.potrs(B, chol)
 -94.2808   -9.3526 -121.8567
  85.0306    7.3517  108.4600
 -17.9665   -0.5677  -22.3295
 339.4752   36.4924  437.1153
 -11.0946    0.4539  -12.3903
[torch.DoubleTensor of size 5x3]

                                                                      [0.0001s] 
th> torch.potrs(torch.cat({B,B}, 2), chol)
bad argument #2 to '?' (Matrix B is rank-deficient at /home/bamos/torch/pkg/torch/lib/TH/generic/THTensorLapack.c:483)
stack traceback:
    [C]: at 0x7f6b7f8d7b90
    [C]: in function 'potrs'
    [string "_RESULT={torch.potrs(torch.cat({B,B}, 2), chol)}"]:1: in main chunk
    [C]: in function 'xpcall'
    /home/bamos/torch/install/share/lua/5.1/trepl/init.lua:650: in function 'repl'
    ...amos/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:199: in main chunk
    [C]: at 0x00405be0

Without the unnecessary check (this PR)

th> torch.potrs(B, chol)
 -94.2808   -9.3526 -121.8567
  85.0306    7.3517  108.4600
 -17.9665   -0.5677  -22.3295
 339.4752   36.4924  437.1153
 -11.0946    0.4539  -12.3903
[torch.DoubleTensor of size 5x3]

                                                                      [0.0004s] 
th> torch.potrs(torch.cat({B,B}, 2), chol)
 -94.2808   -9.3526 -121.8567  -94.2808   -9.3526 -121.8567
  85.0306    7.3517  108.4600   85.0306    7.3517  108.4600
 -17.9665   -0.5677  -22.3295  -17.9665   -0.5677  -22.3295
 339.4752   36.4924  437.1153  339.4752   36.4924  437.1153
 -11.0946    0.4539  -12.3903  -11.0946    0.4539  -12.3903
[torch.DoubleTensor of size 5x6]

[j-wilson]

Was about to make a similar PR; nice catch.

[soumith]

Thanks Brandon!