fix megatron samples

[anoronh4]

Fixed mapping files megatron list for samples (consistent with LIMS and IGO endpoint now)
s_C_000624_N001_d
s_C_000624_P002_d
s_C_001385_M001_d

based on corrections communicated by the PM team. Further information stated below:
s_C_000624_N001_d - originally megatron list had three pairs of fastq from two requests: 06208_D_23 and 07951_C_55 which were believed to come from the same sample. 07951_C_55 was later revised as s_C_000624_N002_d.
s_C_000624_P002_d - one pair of fastqs assigned to this sample originally was later found to belong to s_C_000624_P003_d
s_C_001385_M001_d - one pair of fastqs assigned to this sample originally was later found to belong to s_C_001385_M003_d

all of the re-assigned fastqs were assigned to samples outside of megatron.

[gongyixiao]

• Check with PMs about sample s_C_006624_P001_d, 2 pairs in megatron, 1 pairs in IGO endpoints.
  Update: Megatron is correct. IGO need to fix this in LIMS. Check back later in here: 07951_L_1 possibly missing fq pair in lims LimsRest#131 (comment)
• Change the sample name with _d1 or _d2 to match the LIMS change
• Confirm the rule of not using the re-sequenced samples, and find out in megatron if there is any more samples with multiple request IDs.
  Update: Fixed
• One tumor was paired differently between megatron and Voyager. Need to confirm which one is correct.
  Tumor MegatronNormal VoyagerNormal
  s_C_000624_P001_d s_C_000624_N001_d s_C_000624_N002_d
  Update: stick with Megatron Normal.

[anoronh4]

logging additional changes:

Old CMO ID | New CMO ID
-- | --
s_AML_GCT_2T2 | s_C_006629_T005_d
s_AML_GCT_2T1 | s_C_006629_P002_d
s_AML_GCT_2T3 | s_C_006629_M002_d

These a different mapping was provided by PM but was revised for accuracy between PM and @anoronh4

[anoronh4]

changed s_C_001668_N001_d to s_C_001668_N002_d because IGO changed the sample name on Aug 12