[CK][CK Tile] Fix dram step for KM/KN layouts in V1 pipeline (#5470)

## Motivation

Fix v1 pipeline for KM/KN layouts by passing correct step for dram tile
window.

## Technical Details

- Fix dram step for KM/KN layouts in V1 pipeline
- Disable instances which use more threads than warp size in continous
dim (not supported in ck tile yet)
- Use 1x1 specialization for explicit gemm
- Use two stage for vectorsize =1 and sizeof(datatype) ==2
- remove not needed check sinze GetVectorSizeA/B check if vector size is
fixed

## Test Plan

test_grouped_convnd_bwd_weight_tile

## Test Result

passed locally

## Submission Checklist

- [x] Look over the contributing guidelines at
https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

AICK-966

experimental/grouped_convolution_tile_instances/generate_instances.py

@@ -317,7 +317,7 @@ def parse_bwd_weight_instances(instances, problem_name):
             gemm_params = device_op_name = instance.split("<")[2].split(">")[1].split(",")
             args = [param.split(":")[1].strip() for param in gemm_params]
 
-            spec = "Default"
+            spec = "Filter1x1Stride1Pad0"
             block_size = int(args[0])
 
             mnk_per_block = args[1].split("x")
@@ -450,6 +450,13 @@ def parse_bwd_weight_instances(instances, problem_name):
         if pipeline_version == "V6":
             print(f"Skipping instance {instance_id} with V6 since it's not supported yet.")
             continue
+        if m_per_block > (warp_size * a_scalar_per_vector) or n_per_block > (warp_size * b_scalar_per_vector):
+            print(f"Skipping instance {instance_id} with multiple warps per continous tile dim since it's not supported yet.")
+            continue
+
+        if is_explicit_gemm:
+            if dtype != "float" and c_scalar_per_vector % 2 != 0:
+                is_two_stage_instance = True
 
         conv = ConvInstanceTemplateParams(
             spec,