More fork tests, bug fixes in fork passes and GPU backend

Cargo.lock

Cargo.toml

hercules_cg/src/gpu.rs

hercules_ir/src/fork_join_analysis.rs

hercules_ir/src/ir.rs

hercules_ir/src/loops.rs

hercules_opt/src/float_collections.rs

hercules_opt/src/fork_concat_split.rs

hercules_opt/src/fork_transforms.rs

hercules_opt/src/gcm.rs

hercules_opt/src/lib.rs

hercules_opt/src/unforkify.rs

juno_samples/fork_join_tests/Cargo.toml

juno_samples/fork_join_tests/build.rs

juno_samples/fork_join_tests/src/cpu.sch

juno_samples/fork_join_tests/src/fork_join_tests.jn

juno_samples/fork_join_tests/src/gpu.sch

juno_samples/fork_join_tests/src/main.rs

juno_scheduler/src/pm.rs