[LLVM torture with d8-musl] bcp-1.c failures

[binji]

There were a number of things broken with this test (see bcp-1.c)

• sexpr-wasm was generating the wrong function table entry in (table)
• binaryen is generating the function entry and first argument in the wrong order (see WebAssembly/binaryen#162)
• looks like the test doesn't work in LLVM anyway, opt1 and opt2 seem to be returning 0.

[dschuff]

Closing this as obsolete, and because we are moving the code (see #698). If there is an issue with the build scripts, please file an issue against EMSDK