Refactor ValidationContext and ExecutionContext

[plafer]

Comes out of a comment by @Farhad-Shabani and ensuing sync discussion. We still need to figure out the details, but a main point of confusion was that validate() was in the Context trait; it's not clear from a first look that all the methods in the Context are used in the validate() function. We can also think of different ways to separate the methods into a few different traits if that helps with readability, and doesn't duplicate methods.

Blocked on #221.

Below is the original comment:

Originally posted by @Farhad-Shabani in #221 (comment)

Is it right that we have a tailored validation and execution function for each of the messages like MsgCreateClient, MsgUpdateClient, etc.?
Couldn’t we use the following pattern (as a general concept) and implement them on each message type?

impl validator<Ctx: RouterContext> {
	pub fn validate(&self, ctx: &mut Ctx) -> Result<(), Error>;
}

impl executor<Ctx: RouterContext> {
	pub fn execute(&self, ctx: &mut Ctx) -> Result<(), Error>;
}

// consider it for MsgCreateClient 
impl validator<MockContext> for MsgCreateClient {...}
impl executor<MockContext> for MsgCreateClient {...}

It looks more modular, maintainable, and extensible if it works! (please direct me to the places if I am missing something. tnx)

[plafer]

Suggested grouping of methods: #386 (comment)

[plafer]

We should also keep in mind/revisit the host-based API that was discussed in ADR 5.

[adizere]

We could consider not including this in v1, to avoid ballooning the scope. Let's keep v1 more lean.