[GAP9] unmask FP32 GEMM and SkipConnection paths

Two FP32 kernel tests regressed since the NE16 Linear PR integration:

1. FP32 GEMM/Regular: the new GAP9_NE16GEMMInt32Mapper was prepended to
   the plain Gemm op binding list. It uses the same GEMMParser class as
   the FloatGEMM and GEMMDequant mappers, and Deeploy keys candidate
   bindings by parser class — so listing the NE16Int32 mapper first
   masked the other two. FP32 inputs failed the NE16Int32 type check
   and there was no backtrack to the FloatGEMM mapper.

   Drop GAP9_NE16GEMMInt32Mapper from the Gemm op list (the int8/uint8
   path is already covered by RequantizedGemm with its own NE16 mapper).

2. FP32 Reshape/SkipConnection: MatMulAddMergePass fused MatMul+Add
   into a Gemm with alpha=beta=1 and no transA/transB. The MatMul inputs
   in this test don't share Gemm semantics, so the merge produced
   wrong outputs (16/16 errors). Devel base doesn't include this pass
   in the GAP9 lowering chain and the test passes there. Disable it
   for GAP9 to match.

gvsoc gap9.evk:
  - FP32 GEMM/Regular:           0/1024 errors, 28987 cycles
  - FP32 Reshape/SkipConnection: 0/16 errors,    8343 cycles
  - devel base: 0/1024, 28k-ish; 0/16, 8020 cycles — same shape.

gap9_tiled suite: 96 passed / 1 failed (was 92/5).
Remaining failure (Models/CCT/FP32/CCT_2_32_32_128) also fails at
3b011bb devel base — pre-existing, unrelated to this PR.

Author: runwangdl

Deeploy/Targets/GAP9/Platform.py

@@ -112,7 +112,10 @@
         QuantPatternPass(),
         DequantPatternPass(),
         DequantQuantMergePass(),
-        MatMulAddMergePass(),
+        # MatMulAddMergePass(),  # fuses to Gemm with transA=transB=0 — wrong layout
+        # for MatMul inputs that don't share Gemm semantics; FP32
+        # SkipConnection regressed from 0/16 to 16/16 errors under it.
+        # Leave MatMul and Add separate (matches devel base behavior).
         SkipEmptyConcatPass(),
         SkipUnityRequantPass(previous_op_regex = "Concat", num_inputs = 2),
         SkipUnityRequantPass(previous_op_regex = "Reshape|Transpose", num_inputs = 1),
@@ -142,7 +145,14 @@
     'RequantizedGemm':
         PULPRQSGEMMLayer([GAP9_NE16GEMMMapper, GAP9_MatrixVecMapper, GAP9_TallGEMMMapper, GAP9_GEMMMapper]),
     'Gemm':
-        GEMMLayer([GAP9_NE16GEMMInt32Mapper, GAP9_FloatGEMMMapper, GAP9_GEMMDequantMapper]),
+        # GAP9_NE16GEMMInt32Mapper would also belong here for int8/uint8 Gemm,
+        # but it shares the same GEMMParser class as the other mappers; the
+        # deployer keys candidate-bindings by parser class, so listing it
+        # alongside FloatGEMM / GEMMDequant masks them for FP32 / dequant
+        # paths and the whole graph fails to map. The int8/uint8 path is
+        # already covered by RequantizedGemm above; keep plain Gemm for FP
+        # and dequant flavours only.
+        GEMMLayer([GAP9_FloatGEMMMapper, GAP9_GEMMDequantMapper]),
     'Gelu':
         GELULayer([GAP9_GELUMapper]),
     'LayerNormalization':