Remeta v0.11.1

• New option --skato-collapse-below-aac for remeta gene to collapse variants in SKATO with default 3. Now variants with AAC<3 are collapsed into a burden mask like in regenie. This significantly reduces computation time for large genes like TTN and MUC16.
• New option --cohort-extract COHORT1=FILE1 COHORT2=FILE2 ... for remeta gene to specify an extract list per cohort. When this option is set, only variants in the --cohort-extract extract list for that cohort, and not in the --exclude list, will be included in gene-based tests for that cohort. This overrides the --sources and --extract arguments.
• Fixed a bug in remeta compute-ref-ld where variants with AAF near 1 could have negative variance due to loss of precision when using 32-bit floating point numbers.
• Fixed a bug where tabixable annotation files where not falling back to the rgi index if the tabix index was not available.
• Fixed a bug where using the --chr option with unindexed --htp caused the first line to be dropped.
• The output model column for PVMA of SBAT was updated from BURDEN-SBAT-META_{GENEP_GROUP} to ADD-WGR-BURDEN-SBAT-META_{GENEP_GROUP} to be consistent with other tests.
• Added fallback option to compute standard errors from confidence intervals when missing from the Info column.
• In remeta esma, check that the field INFO in the Info column is a valid double.
• Output more informative error messages when remeta fails to parse a line from an HTP file.
• Added support for 6 column allele frequency file where the sixth column is AN