Speed improvement in caching

[lorenzwalthert]

Don't convert transformers to character. We have version and name, this is sufficient. Goal is also to see if this is reflected in continuous benchmarking.
dccfa

• cache_appyling: 0.095 -> 0.04 (-58%)
• cache_recording: 1.258 -> 1.106 (-12%)
• without_cache: 3.17 -> 3.319 (5%)

6535ba1 (wait: 0.003)

• cache_appyling: 0.095 -> 0.039 (-59%)
• cache_recording: 1.258 -> 1.073 (-15%)
• without_cache: 3.17 -> 3.33 (+5%)

7cac0f1 (wait: 0.004)

• cache_appyling: 0.095 -> 0.047 (-51%)
• cache_recording: 1.258 -> 1.324 (+5%)
• without_cache: 3.17 -> 3.98 (+26%)

75176a6

• cache_appyling: 0.105 -> 0.038 (-64%)
• cache_recording: 1.371 -> 1.085 (-21%)
• without_cache: 3.632 -> 3.555 (-2%)

Major speed improvements are expected for cache_appyling and cache_recording, but without_cache should not change at all but is 5% slower.

[codecov-io]

Codecov Report

Merging #679 into master will decrease coverage by 0.00%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #679      +/-   ##
==========================================
- Coverage   90.43%   90.42%   -0.01%     
==========================================
  Files          47       47              
  Lines        2226     2225       -1     
==========================================
- Hits         2013     2012       -1     
  Misses        213      213              

Impacted Files	Coverage Δ
R/utils-cache.R	100.00% <ø> (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 708b1a1...75176a6. Read the comment docs.