Fix for the TPMQRT routine

Fixes on zgeqrf_qdwh: fix maxmt computing in pztpqrt.c and pztpgqrt. Fixes on cuda_ztpmqrt.c: call CUDA_ztsmqr for TS and TT case for now (TODO: once we have CUDA_zttmqr, we will use it for TT case) : change workspace shift ws

compute/pztpgqrt.c

@@ -58,8 +58,7 @@ void morse_pztpgqrt( int L,
     int ib, minMT;
 
     /* Dimension of the first column */
-    int maxm  = Q2->m - L;
-    int maxmt = (maxm % Q2->mb == 0) ? (maxm / Q2->mb) : (maxm / Q2->mb + 1);
+    int maxmt = (L == Q2->m) ? 1 : ((Q2->m % Q2->mb == 0) ? (Q2->m / Q2->mb) : (Q2->m / Q2->mb + 1));
     int maxmtk;
 
     morse = morse_context_self();