R4R: Fix token printing bug

[cwgoes]

Quoth @ValarDragon:

This stems from a buggy coins.IsValid function. In the IsValid function, https://github.com/cosmos/cosmos-sdk/blob/develop/types/coin.go#L140, it just checks if the coins object is sorted and only has coins with postive value. However when there are multiple tokens, (the default case in the switch statement), the first denom doesn't get its coins value checked for being positive. This means the first denom can be negative or zero. (Recall that its a signed big int >_>, I really wish the uint war would end with everyone being convinced of uints)

This PR adds in desired checks in coins.IsValid() and adds a testcase for a negative first-denom coin.

• Wrote tests
• Updated relevant documentation (docs/)
• Added entries in PENDING.md with issue #
• rereviewed Files changed in the github PR explorer

---

For Admin Use:

• Added appropriate labels to PR (ex. wip, ready-for-review, docs)
• Reviewers Assigned
• Squashed all commits, uses message "Merge pull request #XYZ: [title]" (coding standards)

[codecov]

Codecov Report

Merging #3207 into develop will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff            @@
##           develop    #3207   +/-   ##
========================================
  Coverage    54.83%   54.83%           
========================================
  Files          133      133           
  Lines         9559     9559           
========================================
  Hits          5242     5242           
  Misses        3996     3996           
  Partials       321      321

[alexanderbez]

LGTM 👍

[ValarDragon]

Another thing we should do to mitigate this is make bank.SendCoins ensure all inputs are positive, and that sum_inputs = sum_outputs. We currently only do this on the msg.ValidateBasic, but in order for the keeper to ensure the supply invariant, it must be checked on Keeper.SendCoins as well. (This is another bug contributing to this problem)

[jackzampolin]

Awesome @cwgoes