Add mod supoort in relay.build

[apivovarov]

# in order to compile models we usually call
func, params = relay.frontend.from_<framework>(model_artifact, shape, ...)

# and then we call
graph, lib, params = relay.build(func, target, params=params)

Recently it was a change which changes from_<framework> output to return mod instead of func #3353
But relay.build still supports func as an input. This is why all existing users' scripts to compile models should be updated. All people need to add extra line to their scripts before relay.build

func = mod[mod.entry_func]

In order to avoid all people fixing their compile scripts I think we need to add support for mod input to relay.build.
This PR adds mod input support to relay.build and updates TFLite tutorial.
I can update other tutorials if we agree with this change.

[apivovarov]

@tqchen @jroesch Could you have a look?

[tqchen]

cc @zhiics This seems to be an artifact of incomplete migration to module, I agree with the change but would suggest we move to only require module and add deprecation warnings to function as input

[tqchen]

this is not the best way to check a type in python use

isinstance

[apivovarov]

Ok, I changed it to use isinstance here.

[zhiics]

yeah, I was actually aware of this. But I think this build function is mostly for graphruntime which doesn't actually accept module. Therefore, I don't change it to accept mod because it may confuse people.

[apivovarov]

yeah, I was actually aware of this. But I think this build function is mostly for graphruntime which doesn't actually accept module. Therefore, I don't change it to accept mod because it may confuse people.

build_module has class BuildModule which has method build
Also, build_module itself has helper method build which calls BuildModule.build

In all tutorials to compile models we use helper method build.
What we can do is to change helper method build to accept mod instead of func.
But at the same time keep class method BuildModule.build to still accept func.

[zhiics]

LGTM

[apivovarov]

@tqchen Can you review it one more time?