gpu non-tile works

hercules_cg/src/gpu.rs

hercules_samples/matmul/src/gpu.sch

juno_samples/cava/src/gpu.sch

juno_samples/concat/src/main.rs

juno_samples/implicit_clone/src/main.rs

juno_samples/matmul/src/cpu.sch

juno_samples/matmul/src/gpu.sch

juno_samples/nested_ccp/build.rs

juno_samples/nested_ccp/src/gpu.sch

juno_samples/nested_ccp/src/main.rs

juno_samples/patterns/Cargo.toml

juno_samples/patterns/build.rs

juno_samples/patterns/src/gpu.sch

juno_samples/simple3/build.rs

juno_samples/simple3/src/gpu.sch