[Relay] Refactor Interpreter to treat lowering as IRModule->IRModule rewrite.

[mbs-octoml]

This continues the work outlined in
TECompiler: Rewrite existing compile engine to match updated compiler flow to centralize the TE->TIR lowering logic into a single IRModule->IRModule transformation. This gets about halfway there for the Interpreter:

• Remove direct access to TECompiler from interpreter, and instead call tec::LowerTEExpr when 'preparing' a module and expression for evaluation.
• Make clear there's no phase distinction between create_interpreter and evaluate on the Python side -- both must be prepared together as a single IRModule.
• But in return make sure the result of evaluate on the Python side is a packed func ready to directly apply 'simple' arguments to an already interpreted closure.
• The interpreter builds and caches primitive TIR functions (and their corresponding dynamic shape functions) as packed funcs as they are encountered.
• Clean uses of interpreter for constant folding on the C++ side.

Future work:

• Fold LoweredModule into IRModule so tec::LowerTEExpr is just another pass.
• Get rid of the implicit caching of lowered functions in TECompiler.
• Make calling convention from Relay to TIR explicit, and remove all the function attribute hackery currently needed so the interpreter can correctly invoke lowered functions as it encounters them.
• Make TECompiler private. Though could do this now it will make migrating the VM and AOT uses of CompilerEngine harder.

[mbs-octoml]

@jroesch and @electriclilies for review

[mbs-octoml]

@jroesch PTAL:

• BUG: I was compiling dynamic shape functions for the target even though they must always stay on the cpu. Fixed, but it means a bit more bookkeeping in Interpreter.
• Cleaned up flattening of tuples, both for TupleType/TensorType and runtime::ADT/runtime::NDArray.
  Thanks!

[jroesch]

LGTM, just some stuff we should do as follow up.

[electriclilies]

Thanks for all your hard work Mark! The PR looks good to me so far, I really appreciate the high-quality comments and documentation. I'm still working my way through because it is quite long :)

[electriclilies]

LGTM! Thanks again :)