Remove necessity to keep explicit default policy for MCCFR solvers

[inejc]

These changes are based on discussions in #149. The idea is to remove the need to keep an explicit default policy within MCCFR solvers and thus make them feasible to run on large games where memory constraints are tighter.

The proposed changes allow for usage of:

• TabularPolicy which is the default behavior and the current state
• UniformPolicy for cases where the inferred average CFR policy will only ever be queried with a State instance
• nullptr which will enable to also query the inferred policy with an info state string but move the handling of info state lookup fails to the external caller (empty policy would be returned in that case)

[googlebot]

Thanks for your pull request. It looks like this may be your first contribution to a Google open source project (if not, look below for help). Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please visit https://cla.developers.google.com/ to sign.

Once you've signed (or fixed any issues), please reply here with @googlebot I signed it! and we'll verify it.

---

What to do if you already signed the CLA

Individual signers

• It's possible we don't have your GitHub username or you're using a different email address on your commit. Check your existing CLA data and verify that your email is set on your git commits.

Corporate signers

• Your company has a Point of Contact who decides which employees are authorized to participate. Ask your POC to be added to the group of authorized contributors. If you don't know who your Point of Contact is, direct the Google project maintainer to go/cla#troubleshoot (Public version).
• The email used to register you as an authorized contributor must be the email used for the Git commit. Check your existing CLA data and verify that your email is set on your git commits.
• The email used to register you as an authorized contributor must also be attached to your GitHub account.

ℹ️ Googlers: Go here for more info.

[inejc]

@googlebot I signed it!

[googlebot]

CLAs look good, thanks!

ℹ️ Googlers: Go here for more info.

[lanctot]

This is great, thanks!

I'm making a few (mostly) cosmetic changes. Turns out we don't have to store a tabular policy after all (we have a UniformPolicy object and the best response code does state-based queries already.

Should get merged in Monday's update. (Please don't close the PR.)

[lanctot]

Nevermind, I spoke too soon. They do indeed need to be tabular policies due to assumptions in the best response code (which we should probably tweak eventually...)

[inejc]

Nevermind, I spoke too soon. They do indeed need to be tabular policies due to assumptions in the best response code (which we should probably tweak eventually...)

I see... I suppose this isn't a blocker for the proposed changes?

One minor thing: we should probably rename uniform_policy_ in solvers to default_policy_ since this could, in theory, be something different to uniform now. Please LMK whether I can add commits to this PR or you want to potentially change that later?

[lanctot]

I see... I suppose this isn't a blocker for the proposed changes?

Correct! It's already under internal review, which should be easy to get done today, so will almost surely get merged in Monday's update.

One minor thing: we should probably rename uniform_policy_ in solvers to default_policy_ since this could, in theory, be something different to uniform now. Please LMK whether I can add commits to this PR or you want to potentially change that later?

Yep, that was one of the changes I made :) (renaming to default_policy_) Please don't add commits at this point because I've already imported the PR and the changes would clash with mine.