[CK_TILE] Add conv fwd + bias + clamp example (#3012)

* Implement argument passing to element-wise functions for fwd convolution

* Add files for fwd + bias + clamp example

* Implement Bias

* Implement Clamp

* Elementwise function composition

* Composition unit test

* Implement fwd + bias + clamp example

* Simplify argument passing and composition

* elfunc -> bias_and_clamp

* Rename function to specify example

* Move element-wise function instantiation to kernel

* Make bias a runtime tensor

* No ugly namespace aliasing

* Initialize element-wise function on host

* Remove function initialization helper, simplify Compose initialization

* Remove unintended LSP compatibility patch

* Clean up includes and unused code

* Switch names in cshuffle epilogue

* Move CDElementwise to conv traits

* Re-add required include

* Initialize bias in same way as other tensors

* Better type specification for ds pointer

* Disable 1D convolution

* Add warning for non-group-constant bias

include/ck_tile/ops/epilogue/cshuffle_epilogue.hpp

@@ -8,7 +8,7 @@
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/elementwise/unary_element_wise_operation.hpp"
 
-#include <optional>
+#include <type_traits>
 
 namespace ck_tile {
 
@@ -117,6 +117,10 @@ struct CShuffleEpilogue
     static constexpr index_t MRepeat                       = kMPerBlock / (MPerXdl * MWave);
     static constexpr index_t NRepeat                       = kNPerBlock / (NPerXdl * NWave);
 
+    CDElementwise elfunc_;
+
+    CK_TILE_DEVICE CShuffleEpilogue(CDElementwise elfunc = CDElementwise{}) : elfunc_(elfunc) {};
+
     static_assert(NumDTensor == DsLayout::size(),
                   "The size of DsDataType and DsLayout should be the same");
     /**
@@ -385,7 +389,7 @@ struct CShuffleEpilogue
             generate_tie([&](auto idx) -> const auto& { return ds_tensor[idx]; },
                          number<NumDTensor>{}));
 
-        tile_elementwise_inout_unpack(typename Problem::CDElementwise{}, c_ds_tiles);
+        tile_elementwise_inout_unpack(elfunc_, c_ds_tiles);
     }
 
     template <typename OutDramWindow, typename COutTensor>
@@ -450,7 +454,7 @@ struct CShuffleEpilogue
     CK_TILE_DEVICE auto operator()(ODramWindow& out_dram_window,
                                    const OAccTile& o_acc_tile,
                                    const DsDramWindows& ds_dram_windows,
-                                   void* /*p_smem*/,
+                                   void* /* p_smem */,
                                    const ScaleM& scale_m = {},
                                    const ScaleN& scale_n = {})
     {