LILAC: Off-by-one issue in timing of writing history files

### Brief summary of bug

The LILAC history files are written one timestep later than I would expect.

### General bug information

**CTSM version you are using:** `lilac_cap` branch: `ctsm1.0.dev103-449-g8edcb24a2`

**Does this bug cause significantly incorrect results in the model's science?** No

**Configurations affected:** LILAC

### Details of bug

In this two-day test `LILACSMOKE_Vnuopc_P4x1_D_Ld2.f10_f10_musgs.I2000Ctsm50NwpSpAsRsGs.bishorn_gnu.clm-lilac`, which writes daily LILAC history output, I expected to see two LILAC history files – one at the end of each day. Instead I see just one, one timestep into the second day: `LILACSMOKE_Vnuopc_P4x1_D_Ld2.f10_f10_musgs.I2000Ctsm50NwpSpAsRsGs.bishorn_gnu.clm-lilac.C.20200703_195450_v88bau.clm2.lilac_hi.2000-01-02-01800.nc`.

@mvertens this isn't critical: I'm going to move ahead with bringing this to master without this fix, but this is something we'll want to fix at some point.