build on `approx_log_evidence` introduced in AbstractGPs 0.5.17

[st--]

Deprecates the approx_lml defined in this package in favour of AbstractGPs.approx_log_evidence (see https://github.com/JuliaGaussianProcesses/AbstractGPs.jl/pull/361).

[st--]

hadn't looked so good to the CI :sweat_smile:

[codecov[bot]]

Codecov Report

Patch coverage has no change and project coverage change: -93.01 :warning:

Comparison is base (97675b1) 93.35% compared to head (3f9b152) 0.35%.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #145       +/-   ##
==========================================
- Coverage   93.35%   0.35%   -93.01%     
==========================================
  Files           4       5        +1     
  Lines         286     281        -5     
==========================================
- Hits          267       1      -266     
- Misses         19     280      +261     

Impacted Files	Coverage Δ
src/LaplaceApproximationModule.jl	0.00% <0.00%> (-96.92%)	:arrow_down:
src/SparseVariationalApproximationModule.jl	0.00% <0.00%> (-90.25%)	:arrow_down:
src/TestUtils.jl	0.00% <ø> (-94.12%)	:arrow_down:

... and 2 files with indirect coverage changes

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.

[st--]

@willtebbutt any idea what's going on with the Julia 1 tests? :thinking:

[willtebbutt]

Hmm it looks like something going wrong in ChainRules / Zygote, but I definitely couldn't tell you what 😬