Normalizer generates inconsistent description formats for repeats at c. and r. level with c. duplication input

[ptaschner]

Normalizer input NM_001003187.1:c.877_879dup results in correction to NM_001003187.1:c.876_879[7] due to the repeat expansion, but generates NM_001003187.1:r.(876_879delinsccccccc) for the same repeat at the RNA level (see https://mutalyzer.nl/normalizer/NM_001003187.1:c.877_879dup).
Re-checking NM_001003187.1:r.(876_879delinsccccccc) results in correction to NM_001003187.1:r.876_879[7], which I was expecting in case of consistent application of the repeat rules (see https://mutalyzer.nl/normalizer/NM_001003187.1:r.(876_879delinsccccccc).
The info message correctly mentions that this transcript is not found in the human chromosomal references, because it is a canine transcript.