otp: add tests

[colemickens]

Add tests for OTP functionality. Attempt to prevent future regressions like being discussed in #1006.

Note, this won't pass until one of the proposed fixes is merged:

--- FAIL: TestCalculate (0.00s)
    otp_test.go:21: 
        	Error Trace:	otp_test.go:21
        	            				otp_test.go:26
        	Error:      	Expected nil, but got: &errors.errorString{s:"no totp entry in secret"}
        	Test:       	TestCalculate
    otp_test.go:22: 
        	Error Trace:	otp_test.go:22
        	            				otp_test.go:26
        	Error:      	Expected value not to be nil.
        	Test:       	TestCalculate
FAIL
FAIL	github.com/gopasspw/gopass/pkg/otp	0.002s

Note, a case that I expected to work is not working. It's not listed in the docs, but I was still a bit surprised:

 	check(t, pw, fmt.Sprintf("totp:%s", totpSecret)) 
 	check(t, pw, fmt.Sprintf("totp: %s", totpSecret)) 

[codecov-io]

Codecov Report

Merging #1008 into master will decrease coverage by 0.07%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master    #1008      +/-   ##
==========================================
- Coverage   63.64%   63.57%   -0.07%     
==========================================
  Files         172      173       +1     
  Lines        9932     9965      +33     
==========================================
+ Hits         6321     6335      +14     
- Misses       2824     2842      +18     
- Partials      787      788       +1

Impacted Files	Coverage Δ
pkg/otp/otp.go	42.42% <0%> (ø)

---

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 90bb387...f7dce83. Read the comment docs.

[colemickens]

Rebased and passing. (Not sure I agree that code coverage decreased as a result of this PR...)

cc: @dominikschulz

[dominikschulz]

LGTM. Not sure if I like the check function - would prefer table based tests - but easy to refactor and ok for now.

And yes, the coverage Delta looks wrong.

[colemickens]

@dominikschulz Thanks for the feedback! I went ahead and fixed the structure -- I'd been meaning to switch it to table-tests.

I was going to fix the OTP code checking as well, but it seems like the change in go-kyle/twofactor is not yet sufficient to do this.

[dominikschulz]

LGTM