Optimization for GPU block size determination

This MR optimizes GPU block sizes such that these are always multiples of the hardware's warp (CUDA) or wavefront (HIP) size.

Summarized, this MR

• removes BasicOption GpuOptions.omit_range_check
• removes BasicOption GpuOptions.block_size
• introduces BasicOption GpuOptions.warp_size and implements function for determining default values
• introduces BasicOption assume_warp_aligned_block_size, ensuring the compiler that block sizes match with warp size
• adds new GpuOptions to the data flow of GpuIndexing
• adds algorithm for fitting block size according to iteration space and warp size
• adds fit_block_size and trim_block_size member functions to DynamicBlockSizeLaunchConfiguration for computing block sizes based on a user-defined initial block size and the iteration space
• for assumed alignment: rounds to multiples of warp size when iteration space is unknown to generation time

docs/source/backend/gpu_codegen.md

@@ -72,6 +72,13 @@ These depend on the type of the launch configuration:
 while the `AutomaticLaunchConfiguration` permits no modification and computes grid and block size directly from kernel
 parameters,
 the `ManualLaunchConfiguration` requires the user to manually specifiy both grid and block size.
+The `DynamicBlockSizeLaunchConfiguration` permits the user to set a block size which is used to then dynamically compute
+the grid size. However, the actual block size being used may differ from the user specification. This may be due to
+trimming operations between the original block size and the iteration space to avoid spawning unnecessarily large 
+blocks. In case the `GpuOptions.use_block_size_fitting` option is set, a block fitting algorithm adapts the original 
+block size such that it aligns with the warp size given by the hardware for improved performance. The algorithm 
+`GpuLaunchConfiguration.fit_block_size` incrementally increases the trimmed block size until it is rounded to a multiple 
+of the warp size while considering the maximum amount of threads per block.
 
 The `evaluate` method can only be used from within a Python runtime environment.
 When exporting pystencils CUDA kernels for external use in C++ projects,