Bugfix - wrong HGVSp when --shift_hgvs is enabled

[nakib103]

ENSVAR-6131

In #994 we added a update that will convert a delins to dup. An example can be found in the issue reported in Ensembl/ensembl-vep#500 with example (chr1 20978294 . TTTTATTTATTTA TTTTATTTATTTATTTA). The code change was in _clip_allele function which is used to check after clipping alleles the type of HGVS notation changes. We introduced checking whether type changes to dup before checking for ins.

The _clip_allele function is used by both HGVSc and HGVSp. For HGVSp this change caused a problem as the dup calculation in done after _clip_allele - in _check_for_peptide_duplication function, whereas, for HGVSc it is done before.

So in these code changes I have added condition to not revert delins to dup for HGVSp under _clip_allele.

• also revert back checking = type for only HGVSp (updated in the above mentioned PR without any issue being reported)
• update some comments and refactoring of codes

Test

HGVSp position -

#CHROM	POS	ID	REF	ALT	QUAL	FILTER	INFO	FORMAT	H_NJ-HCC1395-HCC1395
6	41754573	.	C	CCTT	.	.	.	GT:DP:DP4:BQ:SS:AD:DP2:TAR:TIR:TOR:DP50:FDP50:SUBDP50:FT	0/1:76:13,9,28,26:36:2:12:78:12,12:55,55:12,12:80.52:1.92:0:PASS

HGVSc/HGVSp dup gets corrected still -

chr1 20978294 . TTTTATTTATTTA TTTTATTTATTTATTTA
chr1 20978964 . GGTAGGC GGTAGGCGTAGGC

[likhitha-surapaneni]

Tested and LGTM. Thanks @nakib103

[nakib103]

merged to main and postrealeasefix/112