E2 import statement and modules

[Denneisk]

To add to the growing list of Javascript sins on E2, here's another:

As we all know, the E2 namespace is growing every day, and we have a ton of deprecated functions that don't deserve to exist. For new users, these old functions poison the well of how to do things. For old users, they clog up intellisense and the E2 Helper.

Introducing, (pseudo-)modules:

import { callback as fc } from "file"
fc.open("newfileopen.txt", function(a:string) {}, function(b:string) {}) <- ebic
fileOpen("oldfileopen.txt", function ... etc) <- gross

The main difference here is that the module is symbolic. The runtime has no reference to fc, and instead the conversion of fc.open to file.callback.open is done at compile-time. This has actually been a long-standing dream of mine, because I hate that every function in E2 is global.

Immediate benefits:

• Excludes the unnecessary stuff that users will never use.
• Lets users use short names for modules they will use frequently.

Caveats:

• Intellisense will need some work to get it to understand these references. I'm not sure how extensible it is.
• We don't have the infrastructure to support sub-modules or even normal modules.
• Users are going to be confused tf out at first.

Long-term potential:

• We could also extend the syntax to include user E2 files, such as with a prefix e2:myModule.
• We could possibly drop deprecated functions from the global namespace on @strict once adoption is secure.
• Might be able to optimize out some data that'll never be used if a module is not imported, such as never doing the setup for old timers, propcore, etc.

Probably important prerequisites to handle before this:

• Non preprocessed E2 api #2759, definitely