feat: do not deprecate refreshAccessToken

[bcoe]

In #575 @clocked0ne makes a case for it still being useful to be able to call refreshAccessToken in some circumstances; e.g., upstream errors our library does not yet take into account, but might be retriable.

We recently patched a major issue with refreshing tokens, so it's my expectation that the only reason for calling refreshAccessToken would be edge-cases, but I agree with @clocked0ne that it's worth keeping this method available.

fixes #575

[codecov]

Codecov Report

Merging #804 into master will increase coverage by 0.17%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #804      +/-   ##
==========================================
+ Coverage   84.34%   84.52%   +0.17%     
==========================================
  Files          17       17              
  Lines         952      950       -2     
  Branches      211      211              
==========================================
  Hits          803      803              
+ Misses         88       87       -1     
+ Partials       61       60       -1

Impacted Files	Coverage Δ
src/messages.ts	89.47% <ø> (-0.53%)	⬇️
src/auth/oauth2client.ts	82.72% <ø> (-0.06%)	⬇️
src/auth/googleauth.ts	86.25% <0%> (+0.83%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d754771...4b0ec47. Read the comment docs.