Load and visualize large CSV robustly

You are screen-sharing in a HackerRank environment with Python 3, pandas, numpy, seaborn, and matplotlib available.

You are given a single file, data.csv (about 1.5 GB). Its delimiter is unknown (either , or ;) and its encoding is either UTF-8 or latin-1. The columns are:

• id: int
• date: YYYY-MM-DD
• region: str
• spend: float
• clicks: int
• signups: int

Up to 5% of the values may be missing, there can be exact duplicate rows, and some rows have clicks = 0.

Write code to:

1. Detect the delimiter and encoding without loading the full file, then load the data in chunks while keeping peak memory under 1 GB.
2. Drop exact duplicate rows and enforce the column dtypes.
3. Impute missing spend with the median spend within the row's region. Impute missing clicks and signups with 0 only if the row's non-null feature count is at least 4; otherwise drop the row. Justify this rule.
4. Create cpc = spend / clicks using safe division, and winsorize cpc at the 1st and 99th percentiles by region.
5. Produce and save:
   • (a) a scatter plot of spend vs signups with a LOWESS smoothed line and a 95% confidence interval;
   • (b) a boxplot of signups by region, with regions sorted by median;
   • (c) a time-series line of daily total signups.
6. Briefly explain your memory and time complexity choices, and how you would test this code.

Provide runnable, end-to-end code, and state any assumptions explicitly.